HUMAN PERFORMANCE ON
NEGATIVE-SLOPE SCHEDULES OF REINFORCEMENT:
A TEST OF REINFORCEMENT MAXIMIZATION
ERIC A. JACOBS
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1997
I wish to thank the many individuals who helped me complete this dissertation.
Firstly, I wish to recognize my family for all the moral and financial support they have given me over these many years. I am especially indebted to my wife, Angela Rabb Jacobs, for sharing the joys and sacrifices inherent in the pursuit of an academic career. Her unwavering confidence in my abilities has maintained my own. I am grateful to Tim Hackenberg for all the guidance and assistance he has given me. I also wish to acknowledge the other members of my supervisory committee: Marc Branch, Henry Pennypacker, Don Stehouwer, Brian Iwata, and Mark Lewis. I have learned much from the instruction and challenge of these learned scholars. I also wish to thank my fellow students for the many hours of intellectual melee. I particularly wish to acknowledge Manish Vaidya, Cindy Pietras, Dave Stafford, Diana Walker, Karen Anderson and Kevin Jackson for all the conversations, favors, and good times. I am also appreciative of Warren K. Bickel for giving me the opportunity to continue in academics and for his patience during the compilation of this manuscript. Lastly, I wish to express my undying appreciation of E. F. Malagodi for introducing me to the worldview of radical behaviorism. Although I have learned many things from Ed, of most importance has been the value of a good teacher.
TABLE OF CONTENTS
ACKNOWLEDGMENTS ......................................... ii
A B STR A CT ..................................................... v
IN TRODUCTION ............................................... I
EX PERIM EN T I ............................................... 16
M ethod ............................................... 16
Subjects ......................................... 16
A pparatus ......................................... 17
Procedure ......................................... 17
Experimental Conditions ....................... 18
Procedural Irregularities ....................... 22
R esults ............................................... 23
Molar Feedback Functions ............................. 23
Session-by-Session Patterns ............................. 27
W ithin-Session Patterns ............................. 42
IRTs Preceding Point Delivery ....................... 44
D iscussion ............................................... 44
EXPERIM EN T 2 ............................................... 51
M ethod ............................................... 55
Subjects .......................................... 55
A pparatus .......................................... 55
Procedure .......................................... 55
Experimental Conditions ....................... 56
Procedural Irregularities ....................... 58
R esults ............................................... 58
Molar Feedback Functions ............................. 58
Session-by-Session Patterns ............................. 62
W ithin-Session Patterns ............................. 67
IRTs Preceding Point Delivery ....................... 70
D iscussion ............................................... 73
GENERAL DISCUSSION ......................................... 77
Implications for Maximization Accounts ....................... 77
Implications for Matching Accounts ............................. 79
The Role of Verbal Mediation ............................. 82
The Structure of Negative-Slope Schedule Performance ........... 83
SUMMARY AND CONCLUSIONS ............................. 88
REFEREN CES ................... I ........................... 89
BIOGRAPHICAL SKETCH ......................................... 94
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HUMAN PERFORMANCE ON NEGATIVE-SLOPE SCHEDULES OF REINFORCEMENT:
A TEST OF REINFORCEMENT MAXIMIZATION By
ERIC A. JACOBS
Chairman: Timothy D. Hackenberg, Ph.D. Major Department: Psychology
Previous research has demonstrated that the behavior of normal adult humans is more sensitive to temporally remote outcomes than that of nonhumans under laboratory conditions. It has been suggested these between-species differences may be the result of human verbal behavior. The present work attempted to minimize verbal influences by using procedures that are sufficiently subtle to preclude accurate description. In Experiment 1, panel pressing was maintained in five adults by schedules of points exchangeable for money. Following exposure to variable-interval 30-second and linearvariable-interval 30-second schedules, the subjects were exposed to various negative-slope schedules that establish an inverse relationship between response rate and reinforcement rate. Specifically, responses produced points according to an interval schedule and
canceled point deliveries according to a ratio schedule. The way to earn the most points was to respond at a rate of approximately 2 per minute. Overall response rates varied as a function of negative-slope schedule parameters, but overall reinforcement rate was not maximized in any of the negative-slope conditions. Bimodal distributions of interresponse times preceding point delivery and similar within-session patterns of responding were obtained for four of the subjects. In Experiment 2, sensitivity to the inverse relationship between response rate and reinforcement rate was assessed by first exposing two adults to negative-slope schedules, then to yoked-linear-variable-interval schedules that produced the same temporal distribution of point deliveries. Overall reinforcement rate was again not maximized in the negative-slope conditions. Response rates were approximately the same or lower in the yoked-linear-variable-interval condition, suggesting that behavior was insensitive to the contingencies of the negativeslope conditions. The results, however, are difficult to interpret due to order effects and to the absence of response patterning characteristic of steady-state performance on negative-slope schedules. In postexperimental questionnaires, none of the subjects in either experiment were able to quantify the relationship between responding and point delivery. Overall, the results show that when verbal intrusions are minimized, human behavior may be more susceptible to immediate consequences, and more consistent with nonhuman behavior, than has been demonstrated in prior research.
The degree to which behavior is sensitive to short-term versus longer-term
consequences is an area of research with important implications for basic and applied researchers. For the basic researcher, such studies can provide critical information about the time scale over which consequences are functionally related to behavior, which is important in understanding how reinforcement creates and maintains behavioral units. For the applied researcher, such studies may be useful in understanding behavioral problems that stem from excessive control by proximal consequences. For example, people abuse drugs, gamble excessively, incur large amounts of debt, and eat poorly without seeming regard for the temporally deferred consequences of these activities. The ability to demonstrate and manipulate similar "impulsive" behavior patterns in a controlled laboratory setting may greatly increase our understanding of the conditions under which they occur, which may, in turn, make it possible to correct or prevent the occurrence of such behavior outside of the laboratory.
The most widely used procedures to assess control by immediate versus long-term consequences involve mutually exclusive choices between a small immediate reinforcer and a larger delayed reinforcer (Ainslie, 1974; Logue, 1988; Rachlin & Green, 1972). When behavior is governed by the smaller, immediate reinforcer, the performance is said to be "impulsive" or to demonstrate "impulsivity.- Conversely, when behavior is
governed by the larger delayed reinforcer, the performance is said to be "self-controlled" or to demonstrate "self-control."
One of the advantages of such procedures is that they have been used profitably with human and nonhuman subjects alike. In a comparative review of the "self-control" literature, Logue (1988), however, described some general disparities between the performances of humans and nonhurnans on procedures involving recurring choices between an immediate smaller reinforcer and a temporally deferred larger reinforcer. With few exceptions, nonhumans consistently choose the smaller immediate reinforcer, thereby minimizing the delay to reinforcement, whereas verbal adult humans consistently choose the larger delayed reinforcer, thereby maximizing overall reinforcement density. Thus, human performances are said to be "self-controlled," and nonhuman performances are said to be "impulsive." On the face of it, such results suggest that human behavior may be more sensitive than nonhuman behavior to temporally deferred outcomes.
The tendency for human behavior to be more sensitive than nonhuman behavior to long-term outcomes is not limited to the area of self-control, however. Procedures examining choice in situations of diminishing returns have also revealed quantitatively discernible differences between species in sensitivity to temporally remote consequences. Originally introduced by Hodos and Trumbule (1967) with chimpanzees, the procedure involves repeated choices between a fixed schedule, the requirements of which remain constant within a session, and a progressive schedule, the requirements of which are initially lower than the fixed but gradually become more stringent with each reinforcement delivered by that schedule.
The dependent measure of most interest under these procedures is the pattern of
switching from the progressive schedule to the fixed. Under some conditions, short-term consequences (those based solely upon minimizing the delay to the upcoming reinforcer) and long-term consequences (those based upon maximizing overall reinforcement rate) support different choice patterns. The short-term consequences support consistent switching from the progressive schedule to the fixed when the schedule values are equal. Overall reinforcement frequency is maximized, however, by selecting the fixed schedule earlier in the progression, when its schedule requirements greatly exceed those of the progressive schedule.
Subjects consistently switch from the progressive to the fixed schedule in advance of the point at which the schedule values are equal, indicating that behavior is sensitive to consequences beyond the current choice cycle. This effect has been reported with rhesus monkeys, (Hineline & Sodetz, 1987), pigeons, (Hackenberg & Hineline, 1992; Mazur & Vaughan, 1987; Wanchisen, Tatham, & Hineline, 1988) and humans (Hackenberg & Axtell, 1993; Jacobs & Hackenberg, 1996; Wanchisen, Tatham, & 1-ineline. 1992). in addition to the original study with chimpanzees (Hodos & Trumbule, 1967). As with self-control procedures, however, human performances on these procedures are generally better described by accounts based upon the maximization of reinforcement density. whereas the performances of nonhumans are generally better characterized by a delaybased account (Hackenberg & Axtell, 1993; Mazur & Vaughan, 1987). Although the results of these choice experiments have revealed that nonhuman behavior can come under control of temporally removed consequences to some extent, they also suggest that
human behavior may be more sensitive than nonhuman behavior to long-term, overall outcomes.
The between-species disparity in the results of the "self-control" experiments
described above may reflect a fundamental difference between humans and other species in the sensitivity to deferred consequences. It is difficult, however, to establish unequivocally that differences in performance are solely of phylogenic origin. To do so, one would have to rule out at least two other possibilities: (a) that species differences were not due to differences in procedure; or (b) that species differences were not due to differences in preexperimental histories. These will be considered in turn.
Of the many procedural differences between experiments with humans and those with nonhumans, differences in the types of reinforcers may contribute to between-species performance differences. Studies with nonhumans typically use unconditioned remnforcers (e.g., grain presentation to food deprived pigeons), whereas experiments with humans typically use token reinforcers (points exchangeable for money). In "selfcontrol" experiments, for example, humans make repeated choices between a small amount of points exchangeable for money delivered immediately and a larger amount of points delivered after a brief delay. The delay to exchange the points for money is usually fixed and remains the same (i.e., at session's end) regardless of the choice patterns. Humans may consistently prefer the deferred larger amount of points because there is no immediate advantage to obtaining points quickly, as they are not exchanged for consumable reinforcers until later. This hypothesis is supported by the fact that impulsive patterns in humans are more likely with reinforcers that are more immediately
effective, such as escape from white noise (Navarick, 1982; Solnick, Kannenberg, Eckerman, & Waller, 1980) or access to edible reinforcers (Logue & King, 1991). Moreover, recent experiments by Hyten, Madden, and Field (1994) and by Jackson and Hackenberg (1996) have demonstrated that similar performances in people and pigeons, respectively, can be achieved by manipulating exchange delays in "self-control" procedures involving token reinforcers.
Even when procedural differences are minimized, the possibility remains that human operant performance in the laboratory may interact with a subject's prior verbal history (Catania, Mathews, & Shimoff, 1982; Home & Lowe, 1993; Rosenfarb, Newland, Brannon, & Howey, 1992). Humans not only respond to contingencies, but describe those contingencies and how they are related to behavior. Skinner (1969) introduced a distinction between rule-governed and contingency-shaped behavior to help characterize the different ways in which human behavior comes under control of the environment. Contingency-shaped behavior is behavior that is governed by direct experience with contingencies. Rule-governed behavior, on the other hand, is behavior that is under the control of complex antecedent stimuli established through a history of socially mediated reinforcement (Cerutti, 1989; Home & Lowe, 1996; Skinner, 1957). Normal adult humans enter the laboratory with the ability not only to follow rules, but also to derive rules describing the relationship between behavior and consequences. Verbal humans may augment the effects of contingencies with verbal descriptions of those contingencies; as a result, their performances may differ from those of nonverbal subjects.
Like nonverbal behavior, rule-following and rule-generating are classes of behavior established and maintained by consequences. For example, such abilities may be helpful for establishing responding in another member of the verbal community, in occasioning a complicated performance that may have been otherwise forgotten, or in generalizing behavior to contingencies that bear some similarity to those previously experienced. Skinner (1969) suggested that another possible benefit of rule-govemed behavior is to bring behavior into contact with temporally deferred outcomes. According to his view, rules may be helpful in sustaining performances that would normally be curtailed by more proximal consequences. This may account for the between-species differences in performance observed in experiments that place long-term and short-term consequences in opposition.
There is some evidence in the literature to support this position. Adult humans in a "self-control" experiment, for example, reported using maximization strategies that were derived by counting during the intervals (Logue, Pefla-Correal, Rodriguez, & Kabela, 1986). That is, the subjects were able to describe the overall relationship between their choices and the amount of reinforcement. Such descriptions may have functioned as verbal discriminative stimuli that brought their behavior into contact with the deferred consequences.
Additional evidence of the role of verbal behavior in establishing control by deferred consequences comes from a study of age-related differences in self-control by SonugaBarke, Lea, and Webley (1989). In that study, girls ages 4, 6, 9, and 12 made repeated choices between a small reinforcer (one token) and a larger reinforcer (two tokens). The
delay to the small reinforcer was held constant at 10 s, whereas the delay to the large reinforcer was varied from 20 s to 50 s, across conditions. Overall reinforcement maximization was thus contingent upon preferring the smaller, more immediate. reinforcer when the delay to the larger was greater than 20 s. The results indicated developmental differences that corresponded with the subjects' degree of verbal sophistication. The behavior of the four-year-olds was most similar to that of nonhumans. They tended to be either indifferent or to prefer the shorter delay to the smaller reinforcer across all conditions. The six-and nine-year-olds tended to prefer the larger, delayed reinforcer across all conditions, even when doing so decreased the overall amount of tokens obtained. The twelve-year-olds showed increasing preference for the smaller reinforcer as the delay to the large increased, thus approximately maximizing the overall amount of tokens earned. Postsession verbal reports corresponded with nonverbal performances. The six-and nine-years-olds reported preferring the larger, delayed reinforcer regardless of the schedule values, whereas the twelve-year-olds reported preferring whichever alternative would produce the most tokens overall. Although it is difficult to establish the precise role of verbal behavior in generating these performances. the overall pattern of results suggests that verbal behavior may interact with sensitivity to long term outcomes.
There are limits, however, on the extent to which a description can bring behavior into contact with contingencies. Although someone may derive a rule that brings behavior into contact with a given contingency of reinforcement, the rule-governed performance is unlikely to be thoroughly equivalent to a contingency-shaped performance (Skinner.
1969). Training someone to drive a car with a manual transmission provides an example of the use of rule governance to bring behavior under the control of a particular set of contingencies. The novice driver is often given instructions on how to operate the clutch and the accelerator simultaneously. The instructor may say, "Slowly release the clutch while simultaneously depressing the accelerator." Although such instruction is likely to be of some assistance in generating behavior sufficient to move the car, it is highly unlikely that a well executed performance will occur on the first attempt. A fluid departure will occur only after sufficient experience with the contingencies imposed by the car itself. The instructor cannot offer a precise description of appropriate performance based upon his or her everyday interaction with the device. "Slowly" is a rather vague description that is unlikely to occasion smooth, well-executed operation of the clutch on the first attempt. On the other hand, the listener is unlikely to respond appropriately to even the most accurate of instructions under such circumstances. The instructor could take the time and effort to measure precisely the relationship between the operation of the clutch and the accelerator and to refine the instruction given to the pupil to "Lift your left foot off the clutch at a rate of 2 cm per second, while depressing the accelerator with your right foot at a rate of I cm per sec." Without extensive training, however, it is unlikely that the pupil will be able to follow this instruction with any greater degree of success than the original instruction.
It is likely that similar limits of description and interpretation occur when the speaker and listener occupy the same skin. That is, there may be limits to the influence of selfinstruction as well. Expert athletes and artists often cannot describe their own
performances in great detail. Their behavior is shaped and established through an extensive history of contact with the contingencies of the respective activities. Consider the act of driving a golf ball. The distance and the direction the ball moves when hit is determined by subtle aspects of the operation of the club--so subtle that they may elude precise description; one may be left with only vague metaphorical extensions that are of little help in the instruction of others, or oneself.
It is not altogether surprising that human subjects in laboratory experiments can often provide accurate descriptions of the programmed contingencies. Take, for example, the laboratory self-control experiment. Subjects are repeatedly exposed to simple dichotomous choices between alternatives that have rather discriminable outcomes after relatively short delays. These are precisely the conditions under which one might expect accurate verbal descriptions to be brought to bear on nonverbal responding.
Although it is probably not possible to generate a human performance under standard laboratory conditions that does not occasion some verbal behavior with respect to the contingencies, it may be possible to minimize the role of verbal mediation by using contingencies in which the relationship between responding and its consequences is very subtle and difficult to describe accurately. To that end, the present study was undertaken to assess humans' sensitivity to long-term consequences using procedures unlikely to give rise to accurate descriptions. The procedures were patterned after an experiment conducted by Vaughan and Miller (1984), designed to test interpretations of performances based upon maximization of reinforcement density. In that study, pigeons' keypecking was maintained on schedules of reinforcement that established an inverse
relationship between overall response rate and overall reinforcement rate. On this type of schedule, reinforcement density is maximized by responding relatively slowly.
Under these procedures, reinforces are scheduled according to a linear variable interval (LVI) schedule. LVI schedules differ from traditional variable interval (VI) schedules in that upon completion of an interval a reinforcer is added to a "store" of available reinforces and timing of the ensuing interval begins immediately. Reinforcement follows every response when the "store" is positive, and I is subtracted from the "store" following each reinforcement. So, for example, if there were 3 reinforces in the "store," the next 3 responses would be reinforced and the store would be depleted, thus requiring another interval to expire before reinforcement was again available. One way to portray the overall differences between VI and LVI schedules is through the use of molar feedback functions. Molar feedback functions are quantitative descriptions of the relationship between overall rate of responding and overall rate of reinforcement established by a schedule. Figure I shows the feedback functions for both an LVI 30-s schedule (solid line) and for a traditional VI 30-s schedule (dashed line). Whereas the VI function asymptotically approaches the programmed reinforcement rate across the entire range of response rates, the LVI function is constant across all but the lowest response rates. LVI schedules are thus more forgiving of deviations in response rate than traditional VI schedules, in that reinforcement density is maximized for all response rates greater than or equal to the inverse of the schedule value (i.e., 2.0 responses per min in Figure 1).
D 100 90 VI 30s
60 CONSTANT PROBABILITY
0 40 80 120 160 200 240 280 320 360
Responses per Minute
Figure 1. Molar feedback functions for linear and constant-probability VI 30 -s schedules (See text for details).
An inverse relationship between overall response rate and overall reinforcement rate can be established by conjointly subtracting reinforces from the "store" according to a fixed-ratio (FR) schedule. The "store" can be driven into a negative state following periods of relatively high rate responding, thus requiring periods of relatively low rate responding before the value of the "store" is again positive and reinforcement can occur. Figure 2 shows the resulting feedback functions when reinforces are conjointly scheduled by an LVI 30-s schedule and canceled every 100, 50, or 10 responses (from top to bottom, respectively). The functions are negatively sloped across all response rates that are greater than the inverse of the LVI value. For all subtraction ratios, overall reinforcement density is maximized at the point on the function where response rate equals I NI. Beyond this point, faster rates of responding result in decreasing rates of reinforcement until the rate of reinforcement is effectively 0.0 per min. The rate at which the reinforcement rate decreases as a function of response rate is inversely related to the FR value. Note the differences in the slopes of the functions in Figure 2. Reinforcement rate decreases sharply over a narrower range of response rates when reinforcement is canceled every I Oth response than when it is canceled every 50th or every I 00th response.
The feedback function provides a continuum along which performance can be
evaluated with respect to overall reinforcement density. The overall response rate that maximizes overall reinforcement rate will always be the inverse of the LVI schedule value (See Figure 2). Because any significant deviation from this rate will result in a
Linear VI 30 s
120 110 D 100
60 S 50 40 C:
10 NS0 NS NS 50 NS100
0 20 40 60 80 100 120 140 160 180 200 220 Responses per Minute
Figure 2. Molar feedback functions for LVI 30-s and negative slope schedules with subtraction ratios of 10, 50, and 100.
correspondinR decrease in overall reinforcement density, deviations from reinforcement maximization are readily detectable.
To the extent that maximization of overall reinforcement density takes place at the
expense of increases in local delays to reinforcement, the negative slope procedures share some common ground with "self-control" procedures. As response rate decreases on negative slope schedules and approaches the inverse of the LVI schedule value, a greater percentage of the intervals scheduling reinforcement will be shorter than the time between successive responses, or interresponse times (IRTs). As IRT length increases (i.e., responses rate decreases), the probability that the store is positive also increases. Any relative approximation to the rate at which reinforcement density is maximized will therefore incur longer local delays to reinforcement. Thus, as with "self-control" procedures, there is competing control on negative slope schedules by a high probability of immediate reinforcement and increases in overall reinforcement density.
Unlike "self-control" procedures, however, the periodicity of point delivery
established by the schedule may decrease the likelihood that subjects will accurately describe the contingency. Although subjects may be able to report that going "slow" produces more points overall than going "fast," it is unlikely that they will be able to offer a precise quantitative description of the relationship between overall response rate and overall rate of point delivery. If verbal behavior enhances sensitivity to remote consequences, as some research appears to suggest, then perhaps limiting accurate verbal descriptions of the contingencies will also limit nonverbal sensitivity to those contingencies. That is, we may expect to see more "suboptimal" performance (deviations
from maximization) than has been reported in prior studies of human behavior under laboratory conditions.
The present investigation consisted of two experiments. In Experiment 1, panel pressing was maintained in five adult human subjects by schedules of points exchangeable for money. Following exposure to traditional VI 30-s and LVI 30-s schedules, the subjects were exposed to various negative-slope schedules of reinforcement to assess the degree to which their behavior was sensitive to these contingencies. In Experiment 2, sensitivity to the presence of the subtraction contingency was assessed by first exposing subjects to negative slope schedules and then to linearvariable-interval schedules with the same temporal distribution of point deliveries. Performances from the two conditions were then compared to determine if responding was sensitive to the presence of the subtraction contingency. It was hoped that the inverse relationship between responding and point delivery established in the negative slope conditions was sufficiently subtle to preclude accurate description, thereby minimizing verbal mediation.
The present study was also concerned with providing a more detailed picture of
response patterning on negative slope schedules. To that end, daily cumulative response records and IRTs preceding point delivery were collected and analyzed in relation to the contingencies. Thus, apart from the theoretical issues that prompted the use of negative slope schedules to begin with, the study was also aimed at a more thorough understanding of this interesting but seldom used contingency arrangement.
Two male and three female adult humans participated fully in Experiment I in exchange for money. The subjects were recruited via a classified advertisement in a campus newspaper. None of the subjects had previously or was currently enrolled in coursework in behavior analysis or learning theory. The subjects were informed prior to their first session that they would be paid $1.50 per session and that they could receive a bonus by earning points worth $0.05 each. In order to encourage full participation, collection of bonus earnings was contingent upon completion of the study. The subjects were also instructed not to bring personal items (e.g., food, tools, smoking materials, portable radios, or time pieces) into the work space and that violation of this rule would result in dismissal from the study. Overall earnings (including bonuses) ranged from $3.00 to $7.16 per hour (median = $6.00 per hour).
Six additional subjects were recruited for the study but did not participate beyond the initial condition. Five of these subjects withdrew from the study and another was dismissed due to insufficient baseline response rates. The data from these subjects will not be considered here.
The manipulandumn consisted of the right side of a 12.5 cm X 7.5 cm dual response panel, illuminated with red light and mounted centrally on a BRS/Foringer human operant panel (HTC-603). A computer monitor was seated on top of the panel and a keyboard was located in front of the panel. Data were recorded and contingencies controlled by a program compiled in QuickbasicO running on a Compuadd* model 286 computer. Cumulative response records were collected using a Gerbrands* cumulative recorder.
The following instructions were read to the subject prior to the first session and were displayed on the computer monitor prior to the start of each session.
Please read carefully. You have already earned $1.50. In order to receive bonus earnings, you will have to earn points by pressing the red panel. Each point you
earn is worth $0.05. So, for example, if you earn 80 points then your bonus
would be $4.00. You may leave the room at any time in the event of an
emergency. Thank you for your participation.
Sessions were divided into 10-mmn blocks and began when the subject typed his or her subject number on the keyboard and pressed the enter key. Each block was separated by a rest period, during which time the manipulandumn was darkened and no programmed contingencies were in effect. Rest periods were terminated by a response on the computer keyboard. Sessions were comprised of five such blocks for Subject Ill1, and 3
blocks for the other four subjects. Sessions were generally conducted on weekdays at approximately the same time of day.
Table I contains the sequence of conditions and the number of sessions conducted
under each for the five subjects who completed Experiment 1. Conditions were changed when mean response rates and within-session patterns of responding were deemed stable via visual inspection of graphical representations of the data. The following are descriptions of each of the 4 types of experimental conditions.
Variable interval (VI) 30 s. Panel pressing was maintained on a variable- interval (VI) 30-s schedule of point delivery. The VI schedule consisted of 50 intervals generated using the method described by Fleshler and Hoffman (1962). The intervals were randomized and were then presented sequentially both within and across sessions. Upon completion of an interval, the next response would result in point delivery. Point delivery was signaled by a brief tone and by incrementing a counter that was continuously displayed on the computer monitor throughout the block. The next interval in the sequence began timing immediately after point delivery.
Linear variable interval (LVI) 30-s Conditions were the same as in the VI 30 -s
condition except that, upon completion of an interval, a point was "stored" in a bank of available reinforces and the next interval began timing immediately. If a press occurred and the value of the store was positive, one point was delivered and one was subtracted from the store. In the absence of responding, the number of available reinforces in the
Table 1. The sequence of conditions (and the number of sessions conducted under each) for all subjects under variable interval (VI), linear variable interval (LVI), and negative slope (NS) schedules of point delivery.
Sill S888 S999 S211 S521
VI (36) VI (8) VI (10) VI (5) VI (11)
LVI (11) LVI (5) LVI (13) CONJT (8) LVI (7)
NS 326 (35) NS 203 (6) NS 54 (8) VI (5) NS 221 (8)
NS 41 (10) NS 136 (5) NS 30 (9) LVI (5) NS 183 (5) NS 20 (13) NS 100 (6) NS 22 (10) NS 253 (7) NS 156 (8) NS 88 (7) NS 138 (6) NS 139 (17)
NS 74 (5) NS 105 (28) NS 117 (25)
NS 58 (6) NS 33 (29)
NS 47 (7)
NS 10 (9)
store would continue to escalate with the completion of each successive interval. Subjects were not informed of the value of the store, which carried across blocks and sessions, nor of the change in conditions.
Negative slope (NS) FR n. In this condition, points were added to the store according to an LVI 30 s schedule as described above, but every nth response subtracted one point from the store. If points were subtracted more frequently than they were set up by the LVI schedule, the store could be driven into a negative state. Both the value of the store and the position within the FR were carried across sessions. The subtraction ratios for individual subjects are listed in Table 1. These values were based upon the mean response rates from the last five sessions in the preceding condition, and were selected such that the overall rate of point delivery would be reduced by one half if the steadystate response rates of the preceding condition prevailed. In the first negative-slope condition for Subject I 11, for example, the subtraction ratio value was 326, equal to the mean responses per min from the last 5 sessions of the LVI 30 s condition. If this subject continued to respond at 326 responses per min in the NS 326 condition, then one point per minute would be subtracted from the store according to the FR while the LVI 30 s schedule was simultaneously adding 2 points per min to the store. The resulting net reinforcement rate would be approximately one point per minute, or approximately half the steady-state rate of point delivery from the LVI condition.
The subtraction ratios of Subjects 888 and 521 were determined slightly differently.
The initial subtraction ratio was selected such that the overall rate of point delivery would be 0.5 per minute instead of 1.0 per minute. Both of these subjects showed little
sensitivity to the negative slope schedules after several conditions. To produce decreases in response rates in the NS 58 and NS 47 conditions for Subject 888 and in the NS 139 and NS 117 conditions for Subject 52 1, subtraction ratios that reduced the rate of point delivery to approximately 0.0 per minute were used. In the final negative slope (NS 10) condition for Subject 888, the targeted rate of point delivery was -6.3 per minute.
Conjoint VI 30 s (IRT < 2 s):Random ratio (IRT < 2 s). (CONJT VI:RR) For Subject 211, the rate of responding following the initial 5 session exposure to the VI condition was insufficient to progress to the remaining conditions. To rectify this, the subject was exposed to a complex schedule of point delivery in which interresponse times (IRTs) less than 2 s were reinforced according to random ratio (RR) and VI 30-s schedules of point delivery, conjointly. On average, points were delivered following IRTs less than 2 s every 30 s by the VI schedule and after every nth such IRT by the RR schedule, where n equals the random ratio schedule value. The third I 0-min block of the session served as a probe during which the random ratio schedule was suspended and points were scheduled solely by the VI 30-s schedule with the IRT requirement. The random-ratio schedule value was initially set at 30, but was decreased across the first 3 sessions of the condition until response rates increased, indicating schedule control by a response-based contingency. The ratio requirement was then increased gradually over 4 sessions. This manipulation established high rates of responding when the ratio requirement was relatively low and eventually sustained those rates with reinforcement densities comparable to that produced by efficient VI 30-s performance when the ratio requirement was relatively large. At that point, this schedule was discontinued and the VI condition
was implemented. To maintain overall session earnings comparable to the other conditions, sessions were comprised of 3 blocks that terminated after I 0-min or 20 point deliveries, whichever came first.
There were a few deviations from the procedure during the course of the experiment. Subject 2 11 did not respond during the first session. Prior to the second session, the experimenter reread the instructions to the subject and gestured toward the appropriate response panel. For Subjects 111, 888, and 999, the final condition was concluded arbitrarily. For Subjects I I I and 999, the final condition ended due to schedule conflicts. Subject 888 left town during a semester break and did not return to the experiment.
Molar Feedback Functions
Figure 3 shows the mean rate of point delivery versus the mean response rate from the last 5 sessions under all conditions for each subject. These points have been superimposed upon the feedback functions for the respective schedules. The uppermost curve in each graph is the molar feedback function for the LVI condition and the dashed curve below it is the molar feedback function for the VI condition. The diamonds and the triangles in Figure 3 show the data obtained under the LVI and the VI conditions, respectively. Rate of point delivery increased following the transition to the LVI from the VI conditions for all subjects, as indicated in the figure by a slight vertical displacement of the diamonds relative to the positions of the triangles; this result
Figure 3. Mean rate of point delivery versus the mean response rates from the last 5 sessions of the VI 30-s, LVI 30-s, and NS conditions for each subject. The data are superimposed on the molar feedback functions for each condition. Numbers near the abscissae indicate the subtraction ratios of the NS conditions. Note individually scaled axes.
0120 .. ..
NS41 S2N 47 58 74 89 100 NS136 NS203
Q 0 100 200 300 400 500 600 700 0 100 200 300 400
110 VI 30
,oo --A-- VI350s
80 -4 LVI 30 s
60 NS FR x
Ns 22 NS 30 NS 54
Q 25 50 75 100 125
n S211 S521
00NS117 139 156 163 NS 221
NS33 NS105 NS138 NS253
0 100 200 300 400 500 0 1CO 200 300 400 500
Hesponses per Mnut e
is in accord with the programmed changes in the contingencies across the two conditions, and occurred whether response rates increased (Subjects I 11, 211, & 52 1 decreased (Subject 999), or remained the same (Subject 888).
The remaining curves in the graphs of Figure 3 are the molar feedback functions for the negative-slope conditions. The functions are negatively sloped across all response rates greater than 2.0 per minute. Beyond this point, faster rates of responding result in decreasing rates of reinforcement until the rate of point delivery reaches a minimum of 0.0 per min. The rate at which rate of point delivery decreases as a function of response rate is inversely related to the FR value. The filled circles in Figure 3 indicate the obtained rate of point delivery versus the obtained response rate under the negative slope conditions. If reinforcement density was consistently maximized, the points would converge at the apexes of the feedback functions, that is, at 2 responses per min. The response rates of all the subjects across all negative-slope conditions exceeded this value and, therefore, reinforcement rate deviated from the maximum obtainable.
Sensitivity to changes in the contingencies can be assessed by comparing the obtained steady-state values to those encountered earlier in the condition. An appreciation of the situation confronting the subjects following a transition from one condition to the next can be achieved by drawing a vertical line from one of the obtained points in Figure 3 to the abscissa. The point at which this line intersects the feedback function of the subsequent condition will indicate the rate of reinforcement encountered immediately following the transition, assuming similar response rates are obtained. Any steady-state data that are plotted up and to the left of this intersection on the feedback function
indicate a reduction in response rate and a corresponding increase in reinforcement rate. For example, the largest absolute change in response rate occurred with Subject I I I during exposure to the first negative-slope condition (NS 326). A subtraction ratio of 326 was selected such that an overall reinforcement rate of 60 points per hour would be obtained if there were no changes in overall response rate. The rate decreased across the NS 326 condition from 326 responses per minute at the conclusion of the LVI condition to 41 responses per minute at the conclusion of the negative-slope condition. The decrease in response rate resulted in a steady-state reinforcement rate that was comparable to that obtained in the VI condition. The obtained point is plotted up and to the left of the targeted intersection (e.g., 326 responses per minute, 60 points per hour) and, thus, indicates sensitivity to differences in the contingencies across the two conditions.
In Figure 3, complete insensitivity to changes in the FR requirement for reinforcement omission is evidenced by a vertical arrangement of the points from the two conditions. In other words, the response rate would remain the same, as the reinforcement rate changed across conditions according to the different FR values. Across subjects and negativeslope conditions, the only instance of complete insensitivity to changes in the FR subtraction requirement, occurred under NS 100 for Subject 888. The overall rate of responding under NS 100 was similar to that obtained under NS 136. This resulted in a considerable decrease in overall reinforcement rate across the two conditions.
Overall response rate decreased relative to that of the previous condition in the remaining 22 negative-slope conditions across subjects. Although these decreases
indicate sensitivity to the contingencies, the changes in response rate seldom produced increases in the steady-state rates of point delivery across conditions. The steady-state rates of point delivery were lower than those of the previous conditions in 18 of 23 negative-slope conditions across subjects. The five exceptions to this were Conditions NS 89 for Subject 888, NS 22 for Subject 999, NS 105 for Subject 211, and NS 139 and NS 117 for Subject 521. Steady-state rate of point delivery increased across consecutive negative slope conditions only once across subjects (Conditions NS 117 & NS 139 for Subject 521).
The point obtained for Subject 888 under NS 10 deserves comment because it does not fall on the positive region of the molar feedback function for that condition. This subject was exposed to these contingencies for nine sessions and, despite receiving only one point in the first session of the condition, the response rate remained sufficiently high that the rate of point omission exceeded the rate of point availability. By the end of these nine sessions the value of the point "store" was 1424. For the value of the "store" to again become positive, this subject would have had to cease responding entirely for 24 consecutive sessions. The subject left town at this point for a semester break and never returned to the experiment.
Figures 4, 5, & 6 show the mean response rates (filled circles) and mean rates of point delivery (open circles) across sessions for all conditions. Whereas only steady-state data were portrayed in Figure 3, these figures show transition data and trends in response and reinforcement rates.
Figure 4. The mean response rates (filled circles) and mean rates of point delivery (open circles) across sessions for all conditions for Subjects I I I (top panel) and 888 (bottom panel). Note individually scaled axes.
C: 150 0- 125 CiS100
5 15 25 35 45 55 65 75 85 95 105 Sessions
275 V Lo G ;- 00 >- 1c 2.5
0 co C -- 0n It250 (.I,) o/ V.n 0/7 0
2)- 20 z z z 2.0
2 0 0
Q) 125 U
C- 1.00 100 C
5 1015 2025 30 3540 4550 5560 65
For Subject 111, (top panel, Fig. 4) the rate of responding in the VI condition was initially low, but increased across sessions before reaching a steady state of approximately 300 responses per minute. Rate of point delivery varied directly with response rate throughout the VI condition, and asymptotically approached 2.0 points per minute as the response rate increased. Rates of responding and point delivery were comparable to the steady-state rates obtained under the VI condition throughout the LVI condition, although there was somewhat less variation.
Response rates decreased relative to those from the VI and LVI conditions under the negative-slope conditions. At the conclusion of the NS 326 condition, response rates reached a steady state of approximately 41 responses per minute and the resulting rates of point delivery were comparable to those obtained under the VI condition. Under the NS 41 condition, the response rates again decreased, but the decreases were slight in comparison to those of the preceding condition. As a result, the rates of point delivery were lower than those obtained at the conclusion of the NS 326 condition. In the final NS condition for this subject (NS 20), rates of responding were similar to those of the steady-state rates of the preceding condition in all but the final 2 sessions, during which response rates increased slightly. As a result, the rates of point delivery were lower overall than the steady-state rates obtained under the NS 41 condition throughout the condition, and decreased considerably in the last two sessions. Unfortunately, this subject could no longer participate due to a scheduling conflict, and data collection for this subject concluded with this slight increase in response rate.
The bottom panel of Figure 4 shows the mean response rates and the mean rates of
point delivery across sessions for all conditions for Subject 888. Under the VI condition, response rates increased across the first 4 sessions to a condition maximum, then decreased to a steady state of approximately 180 responses per minute. Although there was some between-session variability in the overall rates of point delivery, there were no trends across the condition as a whole. Under the LVI condition, response rates increased slightly relative to the steady state rates of the VI condition but the rates of point delivery were similar to those obtained in the VI condition.
With the exception of NS 100 condition, response rates decreased slightly with each decrease in the subtraction ratio in the negative-slope conditions. Under the NS 100 condition, the overall rates of responding were comparable to that of the preceding condition, but there was an increase in the variability. The overall rates of point delivery also decreased across the NS conditions in 7 of 8 cases. The exception to this pattern occurred in the NS 89 condition, as described earlier.
For Subject 999 (top panel, Fig. 5), response rates increased across the first 5 sessions of the VI condition before decreasing to a steady state of approximately 95 responses per minute. The rates of point delivery varied around a mean of 1.89 points per minute throughout the condition. Response rates decreased throughout the LVI condition and reached a steady state of approximately 54 responses per minute by its conclusion. The rates of point delivery were higher under the LVI condition relative to those of the VI condition, and there was greater between session variability in the rates as well.
Figure 5. The mean response rates (filled circles) and mean rates of point delivery (open circles) across sessions for all conditions for Subjects 999 (top panel) and 211 (bottom panel). Note individually scaled axes.
175 V L <- 0 2.5
(n/ in .1,1 z z
V) 75 /)
a\sI o) o
5 10 15 20 25 30 35 40 45 50 Sessions
Responses O Points
300 V C, V L 6 co N
a) 2.0 a
C 75 .
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Sessions
Under the negative-slope conditions, response rates and rates of point delivery decreased with each successive decrease in the subtraction ratio. Under the NS 54 condition, the response rates decreased slightly relative to the rates of the preceding LVI condition and, as a result, there was a large decrease in rate of point delivery. In addition, between-session variability in rates of point delivery increased under the NS 54 condition. In the subsequent condition (NS 30), there was a slight decrease in response rates relative to the previous condition, although there was considerable overlap in the range of response rates from the two conditions. The rates of point delivery in the NS 30 condition were lower than the rates from the previous condition because the change in response rates across conditions was so slight. Response rates initially decreased under the NS 22 condition and, as a result, the rates of point delivery initially increased. As the condition progressed, however, response rates increased slightly, producing decreases in the rates of point delivery by the conclusion of the condition. Unfortunately, this subject, too, had to leave the experiment due to a scheduling conflict.
The bottom panel of Figure 5 shows mean response rates and mean rates of point delivery across sessions for Subject 211. Due to low response rates in the initial 5 sessions of the VI condition, this subject was exposed to the Conjoint schedule described above. The data from Session 6 were lost due to a programming error. Upon return to the VI condition, response rates decreased slightly, but were sufficient to continue the experiment. Response rates increased under the LVI condition, but the rates of point delivery in the VI and LVI conditions were similar.
This subject was exposed to four negative slope conditions, during which response
rates decreased within and across conditions. Rates of point delivery decreased following the transition from one negative slope condition to the next, but increased within each condition. The largest absolute change in response rates occurred during the third negative slope condition (NS 105); as a result, the steady state rate of point delivery exceeded those of the previous two conditions (NS 138 & NS 253). Response rates decreased further in the final negative slope condition (NS 33) and, following an initial decrease, the rates of point delivery increased to a steady state that was similar to that of the NS 105 condition.
Figure 6 shows the mean response rates and the mean rates of point delivery across sessions for Subject 521. Under the VI condition, response rates increased across sessions, reaching a steady state of approximately 305 responses per minute. With the exception of the first session, the rate of point delivery varied around a mean of 1.80. Response rates increased in the LVI condition, reaching a steady state of approximately 333 responses per minute. The rate of point delivery under the LVI condition was similar to that of the VI condition, but there was somewhat more variability.
Subject 521 was exposed to 5 negative slope conditions. Across the first 3 of these, response rates decreased slightly with each decrease in the subtraction ratio. Rates of point delivery in the first negative-slope condition decreased considerably relative to those obtained in the LVI, then decreased slightly across the two subsequent conditions.
7 '. 2.5
250 4 5
2_ 25 1 5
Q) 175 cn*
0 ,I 0.0
5 10 15 20 25 30 3540 45 50 55 60 65 70 75 80 Sessions
Figure 6. The mean response rates (filled circles) and mean rates of point delivery (open circles) across sessions for all conditions for Subject 521.
The largest absolute decrease in response rate occurred within the final condition (NS 117). The rate of point delivery increased considerably within the condition to reach a level nearly comparable to that of the baseline VI conditions. Within-Session Patterns
With the exception of Subject 999, the within-session response rates were similar
across sub ects. Portions of representative cumulative records from sessions conducted under VI, LVI, and NS conditions are shown in Figure 7 for Subject 211, selected to represent the other subjects. Each panel is the segment of the cumulative record from the second 10 min block of the session. The leftmost segment (Panel 1) is from Session 5 during the first exposure to the VI condition. The slope of the line is shallow and roughly constant across the 10 min block, indicating a low but steady rate of responding. The segment in Panel 2 is from Session 17 during the second exposure to the VI condition (following the Conjoint condition). Note that the slope of the line in this segment is steeper than that of the curve on the left, indicating a higher rate of responding. The next segment is from Session 23 during the LVI condition. The slope of this curve is similar to that from the preceding VI condition, indicating that there was little change in the response rates across the two conditions.
Terminal performance under the negative slope schedules was characterized by alternating periods of high-rate responding and pausing. The development of this response pattern can be seen in the remaining panels, taken from sessions conducted under negative slope conditions. In Panel 4 (NS 253, Session 30), occasional pauses
c PO 0
co c -0
0 0 to
rn V) c C) .0
t > (U
interrupt periods of high-rate responding. In the subsequent condition (NS 138, Session 35), the high-rate patterns were similar, but pausing was more frequent. Panel 6 (NS 105, Session 63) shows response patterning that is characteristic of terminal negative slope schedule performance: pauses alternate with periods of high-rate responding. The highrate response "runs" frequently began with a series of point deliveries as the "store" was depleted and typically continued for many responses thereafter. Occasionally, point delivery occurred in the latter portion of these response runs. This "pause-run" pattern is even more pronounced and consistent in Panel 7 (NS 33, Session 91). Note that the 11pauses" are more frequent and the "runs" are shorter.
The within-session response patterns of Subject 999 differed from those of the other subjects. In the VI condition, the overall rate of responding was lower than those of the other subjects and there was an periodic alternation between accelerating and decelerating response rates, producing a "wavy" cumulative response curve. These general characteristics persisted throughout the experiment. In the LVI condition, the overall rates decreased and momentary fluctuations in response rate became even more pronounced. Under successive negative slope conditions, the overall rate of responding decreased and pauses became more frequent, but the momentary changes in response rate described above persisted. Only in the final negative slope condition was there any evidence of the possible development of a "pause and run" pattern of responding, but even here, this pattern was less pronounced than for the other subjects. Instead, sporadic pausing was often preceded by decelerating response rates and followed by accelerating response rates.
Figure 8. Scatter plots of IRTs preceding point delivery across sessions for Subjects 211 (top panel) and 888 (bottom panel). Note individually scaled and logarithmic axes.
NS 105 NS 33
2 V 0 V L ,t .:
0) 10 0
.. ... .. .
o ~~~ ~ ~ 11 9n.9.*........... ........4
0 10 20 30 40 50 60 70 80 90 100 Sessions
(0 c 0 (7 cc 00 0
o NV) 0 cc LI) t
2 V L L L
'U 102 7 ..' 9.
U) loo 7
10-1 ... .~ .. ..9..
0 ~ 2 10 20 I0 4 0
Figure 9. Scatter plots of IRTs preceding point delivery across sessions for Subjects 521 (top panel) and 999 (bottom panel). Note individually scaled and logarithmic axes.
2 1 Lvi NS 139 NS 1 17
2- co 4
o 102 L :* .:. .
U) 100 ..I* *
sea.* ... ..
10-1~ ~ ~ .. ... ........
2 10____i 20 30 40 50 60 70 80 90 Sessions
2 vL NS-)5 4 N S 3 NS 22
2 1W.; !S*
(1) 10O Ill
0 10 20 30 40 50
IRTs Preceding Point Delivery
The "pause-run" pattern of responding under negative slope conditions apparent in the cumulative records suggested that the IRT preceding reinforcement may be a useful measure to track across conditions. Figures 8 and 9 are scatter plots of the interresponse times preceding point delivery across sessions for all subjects except I 11. (Subject I 11, the first subject run in the experiment, was the subject whose data suggested the present measure.) Consistent with the "pause-run" patterns of the cumulative records, by the conclusion of the negative-slope conditions, the distributions of IRTs preceding point delivery were bimodal for three of these four subjects (888, 211, & 52 1). For Subject 999, although there was an increase in the relative frequency of longer IRTs across the negative slope conditions, the development of a bimodal distribution was not as pronounced.
The general pattern of results suggests that the behavior of all five subjects was
sensitive to the contingencies of the negative slope conditions. This was evidenced by changes in the steady-state rates of responding and point delivery, in the within-session response patterns depicted in cumulative response records, and in the distributions of IRTs preceding point delivery. Steady-state rates of responding decreased relative to those of the preceding condition with each successive decrease in the subtraction ratio in 22 of 23 cases across subjects (See Figure 3). Rates of point delivery, however, were not maximized in any of the negative slope conditions, suggesting that sensitivity to the inverse relationship between response rate and reinforcement rate was limited. The rates
of point delivery tended to increase as response rates decreased within negative slope conditions. Overall, however, the steady-state rates of point delivery generally decreased across successive negative slope conditions. The steady-state rates of point delivery exceeded those of the previous condition in only 5 of 23 negative slope conditions across subjects.
Changes in the within-session patterns of responding also suggest sensitivity to the changes in the reinforcement contingencies. For Subjects I 11, 888, 211, and 52 1, constant, moderate rates of responding established in the VI and LVI conditions gave way to a pattern in which periods of pausing alternated with periods of high-rate responding under the negative slope conditions. For Subject 911, the within-session response patterns were characterized by low rates of responding that fluctuated unsystematically from moment to moment. The overall rates for this subject were, however, lower in the negative slope conditions than in the LVI or VI conditions.
The changes in the temporal distribution of responses depicted in the cumulative records were also reflected in changes in the distributions of IRTs preceding point delivery. The distributions of three of the four subjects (211, 888, & 521) for which this measure was recorded were bimodal, with one cluster of short IRTs (e.g., less than I s) and another cluster of longer IRTs (e.g., greater than 45 s). For Subject 911, the distributions of IRTs preceding point delivery were not bimodal, but the frequency of relatively long IRTs increased under the negative slope conditions.
Some compelling evidence of a limitation in sensitivity to the inverse relationship between overall rate of point delivery and overall response rate comes from sessions in
which unusually long pauses were followed by substantial increases in reinforcement rate. On four occasions across subjects, the end of a I 0-min block was preceded by a particularly lengthy pause (e.g., in excess of 4 min). This occurred twice for Subject 999 (Sessions 3 6 & 3 8) and once each for Subjects I I I and 211 (Sessions 64 and 82, respectively). All subjects were under negative slope conditions at the time. For Subjects I I I and 211, these pauses preceded the conclusion of the final block of the session. Both of these subjects reported having fallen asleep during this period. While they were asleep, the program continued to "store" points as the intervals expired and the "stored" points were carried over to the following session. The subsequent sessions thus began with a series of FR I point deliveries as the "store" was depleted. As a result, the overall rates of point delivery for those sessions were 2.46 and 2. 10 per minute for Subject 111 and 211, respectively (See Figures 4 & 5). For Subject 999, one pause preceded the conclusion of the first 10 min block of Session 3 6 and the other occurred prior to the conclusion of the second block of Session 38. The rest periods following these pauses were also unusually long, suggesting that he, too, had fallen asleep. As with the other subjects, the computer stored points during these pauses and the points were delivered successively at the start of the next block. As a result, the overall rates of point delivery during these sessions were the highest for the condition (See Figure 5). Interestingly, these experiences did not appear to influence immediately the overall rate or pattern of responding in the subsequent session or block.
Sensitivity to the contingencies of the negative slope schedule was apparent earlier in the sequence of conditions for Subject I I I than for the other subjects. The response rates
of Subject I I I decreased substantially during Session 50, the third session of the first negative slope condition (See Figure 4). The "pause and run" response pattern was also observed for the first time in the cumulative record from this session. The other subjects required much longer exposure to the negative slope contingencies before rate decreases of such magnitude were observed, if they occurred at all. These between-subject differences may have been due to a procedural difference that lessened the within-session correlation between rate of responding and point delivery for these subjects. For Subjects 888, 999, 211, and 521 sessions consisted of 3 10-min blocks, whereas, for Subject I 11, sessions consisted of 5 such blocks. Subject I I I experienced each of the 50 intervals comprising the VI 30 s schedule about twice per session. Due to the decrease in session time, the other subjects experienced most intervals once and only repeated 10 intervals per session on average. Therefore, although the session-wide mean interpoint interval varied around 30 s across sessions, the mean for any given session may not have equaled 30 s for these subjects. This resulted in greater variability in overall rate of point delivery under the VI and LVI conditions and lessened the within-session correlation between response rate and point rate under the NS conditions, particularly at relatively large subtraction ratios. The influence of this weaker correlation is apparent in Figures 4, 5, and 6. One might expect that the inverse relationship between overall response rate and overall rate of point delivery would be readily apparent throughout the negative slope conditions in these figures. That is, relatively high rates of responding should produce relatively low rates of point delivery, and vice-versa. Only the data from Subject I 11,
however, show this pattern robustly. The relationship is not as pronounced in the data from the negative-slope conditions for the remaining subjects.
Like the humans in the present study, the performance of nonhumans on negative
slope schedules is not well characterized by reinforcement density maximization. Figure 10 (reprinted from Vaughan, 1982) shows the steady-state response rates from the negative slope conditions in Vaughan and Miller's (1984) experiment superimposed upon the molar feedback functions for the respective schedules. The response rates differed substantially from the inverse of the LVI values, and the reinforcement rates frequently approached 0 per minute. Despite substantial experience with these contingencies, low response rates were not obtained and overall reinforcement density was not maximized. These result have been subsequently replicated with rats (Ettinger, Reid, & Staddon, 1987; Reed & Schachtman, 1989, 199 1).
Comparing the steady-state data presented in Figure 3 to those presented in Figure 10 from the Vaughan and Miller (1984) study suggests that the performances of humans are somewhat similar to those of pigeons. In the Vaughan and Miller (1984) study, each of three groups of pigeons was exposed to a particular subtraction ratio at LVI schedule values of 30, 45, and 90 s. The relevant data for comparison purposes, therefore, are the filled circles plotted on the upper most curves in the panels of Figure 10. These are the data from conditions in which the LVI schedule value was 30 s and the subtraction ratios were 20, 40, and 60, from left to right, respectively. Although interpretation is limited by differences in experimental design (Vaughan and Miller used a between-groups design,
o 40.... ... .
o o 80.... ...
Z 40.... .
0 20 40 60 0 20 40 60 0 20 40 60 RESPONSES PER MINUTE
Figure 10. Reinforcement rates versus response rates of pigeons under NS (filled circles and solid lines) and yoked LVI (open circles and dashed lines) conditions from Vaughan and Miller (1984). The data have been plotted on the molar feedback functions for the respective conditions. From "Choice and the Rescorla-Wagner model." by W. Vaughan, Jr., 1982, in M. L. Commons, R. J. Herrnstein, & H. Rachlin (Eds.), Quantitative analyses of behavior: Vol. 2. Matching and maximizing accounts, (pp. 263 -279), Cambridge, MA: Ballinger. Copyright 1982 by Lawrence Erlbaum Associates, Inc. Reprinted by permission.
the present study a within-subject design), the data in Figure 10 suggest that rates of responding and reinforcement were inversely related to the subtraction ratio value. This was frequently observed with the human subjects as well (See Figure 3).
There were two instances in the present study, however, that were not consistent with this pattern. In the NS 105 condition for S21 I and the NS 117 condition for S52 1, response rates decreased sufficiently to produce substantial increases in the rates of point delivery relative to those of conditions with larger subtraction ratios (See Figure 3). These deviations may have been driven by shifts in the within-session patterns of responding. The large absolute changes in rates of point delivery within these negative slope conditions were correlated with the development of the "pause and run" pattern of responding. Similarly, although it was the first negative slope condition for Subject I 11, large absolute changes in rates of point delivery occurred in the NS 326 condition, where the "pause and run" pattern first developed.
Although cumulative response records were not presented for the pigeons in the
Vaughan and Miller (1984) study, it is less likely that such patterns would have develop because a minimum 5 s interreinforcement interval was imposed. That is, if there was more than one reinforcer "stored", pecks within 5 s after reinforcer presentation would not produce another reinforcer. This contingency was omitted from the present study to enhance the inverse relationship between response rates and reinforcement rates, thereby providing a more stringent test of reinforcement density maximization.
That human subjects consistently failed to maximize reinforcement density under the negative slope schedules in Experiment I raises the question of whether or not their behavior was sensitive to the inverse relationship between response and reinforcement rates. Instead, their performances may have been governed solely by the obtained overall rate of reinforcement. That is, it is possible that the pattern of reinforcement generated under the negative slope schedules influenced the overall rate and pattern of responding in a manner similar to that of a variable interval schedule with the same temporal distribution of reinforcement; the subtraction ratio may have exerted only an indirect influence by determining that distribution.
Hermstein (1970) introduced a hyperbolic functional relation to account for steadystate response rates on VI schedules. It is based upon the premise of the matching law which states that the relative rate of responding equals the relative rate of reinforcement produced by that behavior. The expression takes the general form of B = L(r/(r + io)) (1)
where B equals response rate, r equals the obtained rate of reinforcement produced by B, k equals the maximum value of B, and ro equals the reinforcement rate corresponding to 1/2 B (Bradshaw, et al., 1976); lo is theoretically equal to the rate of unscheduled reinforcement produced by behavior other than B. With the aid of the two free 51
parameters, Hermstein (1970) was able to fit this curve to data obtained by Catania and Reynolds (1968) with pigeons responding on a variety of VI schedules.
The data from the negative slope conditions in the Vaughan and Miller (1984)
experiment suggest that overall response rate might match overall reinforcement rate. The filled circles in Figure 10 show the obtained response rates versus the obtained reinforcement rates from negative slope conditions. Although none of the pigeons was exposed to enough conditions to discern a hyperbolic relationship, response rates varied directly with reinforcement rates for 6 of the 9 birds, an outcome that is consistent with accounts based upon relative reinforcement rate (Hermstein, 1970; Prelec, 1982). This outcome suggests that the pigeons' behavior might not have been sensitive to the subtraction ratio of the negative slope schedule.
The overall pattern of results from Experiment 1, however, indicates that human performance on similar negative-slope schedules was at least partially sensitive to the subtraction contingency. With the possible exception of Subject 888, the distributions of steady-state data in Figure 3 are not well characterized by hyperbolic functions. In the coordinates of Figure 3, the function described above would be plotted as a hyperbola that is concave upward, intersecting the ordinate at 0 responses per minute. The function would asymptotically approach some maximum rate of responding along the abscissa. According to the equation above, relatively higher rates of reinforcement should result in relatively higher rates of responding. Although this relationship holds for 18 of 23 negative-slope conditions across subjects, fairly pronounced decreases in response rate were accompanied by increases in reinforcement rate in four of the five exceptions. The
data from these conditions are not in accord with a straightforward application of Equation 1.
Experiment 2 was conducted to assess explicitly sensitivity to the relationship between rates of responding and reinforcement under negative slope schedules with human subjects. One way to assess sensitivity to such a relationship is to compare performance under these conditions to performance under a variable-interval schedule with the same temporal distribution of reinforcement. For example, subjects might first be exposed to a negative-slope schedule, then to a linear variable-interval schedule in which the reinforcement rate is yoked to that obtained in the preceding negative-slope condition. If the performances in the two conditions are similar, then it suggests that the overall rate of reinforcement is the relevant controlling factor and that responding is insensitive to the subtraction contingency. If response rates are consistently lower under the negative slope schedule than under the variable interval schedule, then it implies that the subjects' behavior is indeed sensitive to the inverse relationship between responding and reinforcement.
The results of such a manipulation can be seen in Figure 10. In the Vaughan and Miller (1984) study, the negative slope conditions alternated with LVI conditions in which the overall reinforcement rates equaled those of the preceding negative-slope conditions. The data from the LVI conditions (open circles) have been superimposed upon the molar feedback functions for those schedules (dashed lines) in the figure. Although not a particularly robust effect, the response rates tended to be higher in the
LVI conditions than in the associated negative slope conditions, indicating some degree of sensitivity to the negative slope contingencies.
The tendency for higher response rates under the yoked LVI schedules than under negative slope schedules with comparable reinforcement rates has been replicated in a series of experiments by Reed and Schachtman (1989, 199 1) with rats as experimental subjects. In two of these experiments between-subject designs were used. Animals were paired according to similarities in their baseline performances. One rat in each pair was then arbitrarily assigned the master subject, exposed to a negative slope schedule of food delivery, and the other was assigned the yoked subject, exposed to an LVI schedule in which the reinforcement rates were yoked to those experienced by the master subject. In one study, reinforcement was yoked solely in terms of the overall rate, but in the other, the temporal distribution of the reinforcement was matched across the two subjects (Reed & Schachtman, 1989 & 1991, respectively). In the latter study, the roles of the subjects were later reversed, such that each subject had experience as both a master subject and a yoked subject. In both experiments, rats in the yoked groups tended to respond at slightly higher rates than the master rats that experienced the negative slope schedules; however, between subject differences in performance made interpretation of the results tenuous, at best.
In another experiment by Reed and Schachtman (199 1), the comparison between performances under negative-slope and yoked LVI schedules was accomplished in the context of a within-subject design. Food was delivered according to a negative slope schedule in one component of a multiple schedule of reinforcement and according to a
yoked LVI in another. The component schedules were presented in strict alternation following each food delivery. Response rates were higher in the yoked components than in the negative slope components for all subjects, suggesting that their behavior was indeed sensitive to the inverse relationship between response rate and reinforcement rate in the negative slope components.
Experiment 2 extended the use of the yoked-control procedure to human subjects to assess sensitivity to the subtraction contingency. Using a reversal design, each subject was exposed to negative-slope and yoked-LVI schedules of points exchangeable for money. The interpoint intervals obtained under the negative slope condition were later presented in the same sequence in the yoked LVI condition. VI 30 s and LVI 30 s conditions served as the baseline conditions to which subjects were exposed prior to both the negative-slope and the yoked LVI conditions.
Two male adult humans participated fully in Experiment 2. All of the conditions of their participation were the same as those described above for the subjects in Experiment I .
The apparatus was the same as that used in Experiment 1. Procedure
The instructions described above for Experiment I were read to each subject prior the first session and were displayed on the computer monitor before the start of each session.
Sessions consisted of three 10-min blocks separated by rest periods that were terminated with a response on the computer keyboard. As in Experiment 1, the subject initiated the start of the session by entering his subject number. Single sessions were conducted on weekdays at approximately the same time of day. Experimental conditions
Table 2 contains the sequence of conditions and the number of sessions conducted
under each for the two subjects. Conditions were changed when mean response rates and within-session patterns of responding were deemed stable via visual inspection of graphical representations of the data. The following are descriptions of each of the 4 experimental conditions.
Variable interval (VI) 30 s and linear variable interval (LVI) 30 s. The conditions were the same as those described above for Experiment 1.
Negative slope (NS) FR n. The contingencies of the negative slope conditions were the same as those described in Experiment 1. As before, the determination was based upon the mean response rates from the final 5 sessions of the LVI condition; the value was selected such that the overall rate of point delivery would equal 0.5 points per min if the response rate of the LVI condition prevailed.
Yoked Linear Variable Interval (Yoked LVI). A linear variable interval schedule was in effect during this condition. The distribution of intervals used to schedule point delivery
Table 2. The sequence of conditions (and the number of sessions conducted under each) for both subjects in Experiment 2 under variable interval (VI), linear variable interval (LVI), negative slope (NS), and yoked LVI schedules of point delivery.
VI (24) VI (29)
LVI (16) LVI (9)
NS 17(11) NS 10(13)
VI (9) VI (10)
LVI (4) VI LH 2s (10)
NS 16 (12) VI (8)
VI (14) LVI (7)
LVI (9) Yoked LVI (13)
Yoked LVI (12)
was composed of the interpoint intervals obtained under the NS n conditions for each subject, presented in the same sequence in which the points were delivered. Procedural irregularities
For Subject 711, the response rates during the initial 5-session exposure to the VI
condition were very low. In an attempt to increase response rates, relatively shorter IRTs were differentially reinforced manually in Sessions 5 through 10. Also, due to a programming error, the interpoint intervals obtained from Subject 711 under the NS 17 condition were inaccurate, which necessitated an additional negative-slope condition. In an attempt to recover VI response rates before proceeding to the yoked LVI condition, Subject 911 was exposed to a VI 30 s with a 2-s limited hold. The contingencies in effect during this condition were similar to those of the VI 30 s condition except that upon completion of an interval, reinforcement was made available for only 2 s. If a response did not occur within 2 s from the completion of an interval the point delivery was canceled and the next interval began timing.
Molar Feedback Functions
Figure I I shows the mean rate of point delivery versus the mean response rate from the last 5 sessions under all conditions for Subjects 711 (top panel) and 911 (bottom panel). As in Figure 3 from Experiment 1, these points have been superimposed upon the feedback ftmctions for the respective schedules. Due to the reversal design employed in Experiment 2, the subjects received multiple exposures to the VI and LVI conditions. For clarity, data from similar conditions share the same symbol and the points have been
Figure 11. Mean rate of point delivery versus the mean response rate from the last 5 sessions under al I conditions for Subjects 711 (top panel) and 911 (bottom panel). The data have been superimposed upon the molar feedback functions for the respective schedules. Data from similar conditions share the same symbol and the points have been numbered according to the sequence in which were implemented (See Table 2 and the text for details). Note individually scaled axes.
1 0. .... .'.
100 1 A
0 NS16 NS17
0 5 10 15 20 25 30 35
C S91 1
0 100 A
40 --A-- VI30s
30 LVI 30 s
0 VI 30 s LH
0 NS1O0 5 10 15 20 25 30 35 40 45 50 55
Responses per Minute
numbered according to the sequence in which were implemented (see Table 2). Sensitivity to changes in the contingencies can be assessed by comparing the relative heights and left-right positioning of points across successive conditions. Consider, for example, the relative positions of the points labeled I and 2 for Subject 711. These data are from the first exposures to the VI and LVI conditions, respectively. Point 2 lies up and to right of Point 1, indicating that rates of responding and point delivery increased in the LVI condition.
For Subject 711, the steady-state rates of responding and point delivery were consistently lower in the VI conditions (Points 1, 4, & 7) than in the ensuing LVI conditions (Points 2, 5, & 8, respectively). The steady-state rates of responding and point delivery in the NS conditions (Points 3 & 6) were lower than those of the immediately preceding LVI conditions (Points 5 & 8), indicating sensitivity to the negative slope contingencies. The steady-state response rates in the yoked LVI condition (Point 9) were substantially lower than the steady-state rates from the preceding LVI condition (Point 8) and slightly lower than the negative slope condition in which the interpoint intervals were obtained (Point 6).
For Subject 911 (bottom panel), steady-state rates of responding were higher in the first VI condition (Point 1) than in the subsequent LVI condition (Point 2). In the negative-slope condition (Point 3), steady-state rates of responding and point delivery decreased. Upon return to the VI condition (Point 4), steady-state rates of responding increased only slightly. To rectify this, a limited-hold on the availability of points scheduled by the VI 30 s schedule was implemented. Although response rates increased
within this condition (Point 5, See also Figure 12 below), the rates decreased again once the limited hold was discontinued (Point 6). Steady-state response rates decreased slightly in the subsequent LVI condition (Point 7). As with Subject 711, the steady-state response rates in the yoked LVI condition (Point 8) were lower than the steady-state rates from the preceding LVI condition (Point 7) and from the negative slope condition in which the interpoint intervals were obtained (Point 3). Session-by-Session Patterns
Figure 12 shows the mean response rates (filled circles) and mean rates of point
delivery (open circles) across sessions for all conditions for both subjects. Whereas only steady-state data were portrayed in Figure 11, this figure shows transition data and trends in response and reinforcement rates within each condition. For Subject 711 (top panel), response rates were initially very low in the first VI condition. In an attempt to increase response rates, relatively shorter IRTs were differentially reinforced manually in Sessions
5 through 10. Rates of responding and point delivery increased during these sessions. The effect was short lived, however, as response rates decreased throughout the remainder of the condition before reaching a steady state of approximately 20 per minute.
The interpoint intervals used to schedule the point deliveries in the yoked LVI
condition were obtained in the NS 16 condition, because the interpoint intervals obtained in the preceding NS 17 condition were inaccurate due to a programming error. Rates of responding and point delivery decreased in this condition relative to those of the immediately preceding LVI condition. In the ensuing VI and LVI conditions, the rates of
Figure 12. The mean response rates (filled circles) and mean rates of point delivery (open circles) across sessions for all conditions for Subjects 711 (top panel) and 911 (bottom panel). Note individually scaled axes.
100 V VLO V w
J 75 2.5
0 0. 5 U
0 5 1 5 25 35 4 5 55 65 75 55 95 105 0.
0 Resp-onses o Points V0 Lv V VLH V L OE 2.5
5 1 25 3 45 55 65 75 85 95 105
responding and point delivery increased to levels comparable to those of the previous exposures, indicating a recovery of baseline performance.
When the yoked LVI condition was implemented, response rates were slightly lower than the rates in the preceding LVI condition, and approximately equal to the rates in the NS 16 condition by the conclusion. The mean response rate from the last 5 sessions under the yoked LVI condition was 42% of the mean response rate from the last 5 sessions of the immediately preceding LVI condition, whereas the mean response rate from the last 5 sessions of the NS 16 condition was 58% of the immediately preceding LVI condition.
Note the high degree of correspondence between the rates of point delivery in the NS 16 and yoked LVI conditions, which is in accord with the programmed contingencies. The only deviations from exact accordance occurred in Sessions 107 and 108. The store was positive at the conclusion of Session 107, with a point carried over to Session 108. As a result, the overall rate of point delivery was slightly lower in Session 107 than in Session 74 (the session to which it was yoked) and slightly higher in Session 108 than in Session 75.
The bottom panel of Figure 12 shows the mean response rates (filled circles) and
mean rates of point delivery (open circles) across sessions for all conditions for Subject 911. Early in the initial exposure to the VI condition, response rates were low, but then increased across sessions. Rates of point delivery also increased gradually across the condition. In the subsequent LVI condition, however, response rates decreased considerably while rates of point delivery increased slightly.
In the NS 10 condition, response rates increased across the first three sessions and
remained higher than the steady state rates of the preceding LVI condition until Session 43. As a result, no points were delivered in 5 out of the first 7 sessions of the condition. By Session 46, however, response rates had decreased sufficiently to allow the "store" to again become positive. Variability in response rates decreased in the remainder of the condition, with rates of point delivery reaching a steady state of 1.08 per minute.
When the VI conditions were reinstated, the rates of responding and point delivery
increased only slightly and were much lower than the rates obtained in the first exposure. In an attempt to recover baseline rates of responding, a 2-s limited hold was implemented. By the conclusion of this condition response rates increased and were comparable to the rates obtained during the initial exposure to the VI condition, but when the VI condition was reinstated, the rates of responding again decreased. Due to time constraints, no additional manipulations to increase response rates were implemented and the experiment progressed, despite the failure to recover baseline response rates completely.
In the first five sessions of the yoked LVI condition, few points were delivered and response rates increased slightly relative to those at the conclusion of the preceding LVI condition. In Session 96, there was a period of high rate responding during the third block, resulting in a considerable increase in session-wide response rate despite an absence of point deliveries. Once the yoked LVI condition had reached the point in the sequence of sessions at which point delivery occurred again, the response rates had decreased. By the conclusion of the condition, response rates were somewhat less than
the rates from the corresponding negative slope sessions. The mean response rate from the last 5 sessions under the yoked LVI condition was 47% of the mean response rate from the last 5 sessions of the immediately preceding LVI condition, whereas the mean response rate from the last 5 sessions of the NS 16 condition was 55% of the immediately preceding LVI condition.
Sections of representative cumulative records from sessions conducted under NS and yoked LVI conditions are shown in Figure 13 for Subject 711 (top panels) and 911 (bottom panels). Each panel is the segment from the second I 0-min block of the session. Under the negative slope conditions, neither subject showed the "pause-run" pattern prevalent for three of the four subjects in Experiment 1. Although relatively long pauses that concluded with point delivery were fairly frequent, they were often separated by periods in which responding occurred at low rates that fluctuated slightly from moment to moment. Under the yoked LVI condition, overall response rates decreased for both subjects, as indicated by the shallower slope of the records on the right. The withinsession patterns under the yoked LVI conditions differed for the two subjects, however. For Subject 911, responding occurred at a low, but fairly constant rate throughout the session. For Subject 711, the characteristics of the within-session patterns were similar across the two conditions. Relatively long pauses ending in point delivery became more frequent in the yoked LVI condition, but they were still separated by periods of fluctuating low rate responding. The response patterns of both subjects occasionally produced changes in the temporal distribution of point deliveries.
Figure 13. Portions of representative cumulative response records from sessions conducted under NS and yoked LVI conditions for Subjects 711 and 911. Each panel is the segment of the cumulative record from the second 10 min block of the session.
Session 76 Session 111
Lo 300 i)
I ___ ___ ___ ___ __O 5 10 0 5 10
NS 10 YOKED
Session 50 Session 102
(D 400 U)
I I I II
0 5 10 0 5 10
IRTs Preceding Point Delivery
Scatter plots of the IRTs preceding point delivery across sessions for Subjects 711 (top panel) and 911 (bottom panel) are shown in Figure 14. The relevant comparisons in this figure are between the distributions from the yoked LVI conditions and the distributions from the negative slope conditions in which interpoint intervals were obtained (NS 16 and N S 10, for Subjects 711 and 911, respectively). The distributions of IRTs from yoked LVI conditions differ from those of the corresponding negative slope condition for both subjects. For Subject 711, the distribution of IRTs was bimodal in the NS 16 condition. Interestingly, this distribution of IRTs was not readily apparent in the cumulative records for this subject. There was a cluster of IRTs centered around 17 s and another around 0.2 s, and no IRTs were longer than 70 s. In the yoked LVI condition, the distributions maintained a bimodal appearance, but the IRTs were more widely dispersed about the modes. There was a greater frequency of IRTs within the range of 0.5 to 5 s in the yoked LVI condition than in the NS 16 condition and the frequency of relative long IRTs increased, with many exceeding 70 s.
For Subject 911, the distribution of IRTs under the NS 10 condition was not bimodal. At the conclusion of the condition, the IRTs were most densely clustered around 17 s and ranged from approximately 0.25 s to 40 s. In the yoked LVI condition, the frequency of IRTs less than 2 s was considerably less, and the frequency of IRTs exceeding 25 s was higher than in the negative slope condition. Because points were not delivered in some of the early sessions of the NS 10 condition, there were no IRTs preceding point delivery in these sessions or in the corresponding yoked LVI sessions.
Figure 14. Scatter plots of IRTs preceding point delivery across sessions for Subjects 711 (top panel) and 911 (bottom panel). Note individually scaled and logarithmic axes.
102 Z.01 I! ..I
n2 ** 4
C7. 0 3 .. .
2 10n 3*i 40 50* 60 77 8** ICO 1
___________________ _________ I______ _______________ _V_102
>3 102 (
11-- ils '.*. :1
i L : *0 t I .
7* 3 s ~ *oz.8 r
.) 2 I
Q 10-i 1
10 20 770 40 0 o ess ions
The results from the negative slope conditions of Experiment 2 replicate those of Experiment 1. Response rates were lower in the negative slope conditions than in the initial VI baseline conditions, but were not low enough to maximize overall rate of point delivery. Unlike rats (Reed & Schachtman, 1989; 1991) and pigeons (Vaughan, 1982), however, the response rates of both subjects were slightly lower in the yoked LVI condition than in the corresponding negative-slope condition (see Figures I I & 12). At first glance this might suggest that human behavior is not sensitive to the inverse relationship between rate of responding and rate of point delivery. The results, however, are difficult to interpret because the schedule performances of these two subjects differed from those of the subjects in Experiment 1, suggesting that schedule control may have been partial or incomplete. Specifically, the response rates of these subjects were much lower in the VI and LVI conditions than those of the subjects in the previous experiment. Also, under negative-slope conditions, the distinctive "pause and run" pattern evident in the cumulative records of four of five subjects in Experiment I was not apparent for either subject (see Figure 13) and bimodal distributions of IRTs preceding point delivery were obtained for Subject 711 only (see Figure 14). If these characteristics of performance are typical of terminal performance on negative slope schedules, then perhaps the subjects in Experiment 2 did not receive sufficient exposure to the negative slope contingencies.
The reversal design employed in Experiment 2 may also have produced troublesome history effects that complicate interpretation of the results. For Subject 911, response
rates decreased in the first exposure to the LVI condition and, following exposure to the negative slope contingency, it was not possible to recover baseline levels of performance. Exposure to these conditions possibly attenuated sensitivity to additional manipulations of the contingencies. These results may not be too surprising given the prevalence of human behavioral persistence under laboratory conditions (Weiner, 1969, 1970, 1983). As Weiner (1983) has noted, human response patterns tend to persist unless "adverse reinforcement consequences induce change" (p. 521). For Subject 911, the transition from the negative slope condition to the VI condition was met with an increase in rate of reinforcement. These are precisely the circumstances under which one might expect to see persistence of the response patterns established in the earlier condition. When a 2-s limited hold on the availability of points scheduled by the VI was imposed, reinforcement rates decreased initially and then increased as response rate increased. The initial decrease in reinforcement rate was sufficient to induce a change in responding. Rates of point delivery increased when the limited hold was discontinued and remained elevated despite decreases in response rates. Under the yoked LVI condition, previously established patterns of responding were likely to persist because there were no decreases in reinforcement rates to drive changes in response rates.
In a postexperiment questionnaire, Subject 911 reported a contingency description and a performance strategy that were broadly consistent with the contingencies of the negative-slope condition and with his performance. In response to the query; "Did pressing the red panel influence point delivery? If so, please describe how as best you can."
Subject 911 wrote;
". .. It seemed to me points were only available if you pressed the red button. But within the tests, it seemed to me that if you pressed the red button too often points were more elusive, which led me to a strategy of restraint in how often I pushed the button."
Although the origin and causal status of this verbal report are dubious, it is consistent with his nonverbal performance, and suggests that at least some aspect of the subject's behavior was sensitive to the negative-slope contingencies. The behavior produced by the negative slope condition (verbal and nonverbal) may have also been occasioned in the yoked LVI condition by the sudden decrease in rate of point delivery. It is then possible that the contingencies of the yoked LVI condition may have sustained these patterns of behavior. That is, Subject 911 may have been sensitive to presence of the subtraction contingency in the negative slope condition, but not sensitive to the absence of the contingency in the yoked LVI condition. Performance in the yoked LVI condition may have been an instance of what Shimoff, Catania, and Mathews (1986) referred to as "pseudosensitivity" to the contingencies--rule-governed response patterns in accord with the contingencies. In the present case, the response rates on the yoked LVI may have been produced directly by the rate of point delivery. Alternatively, they may have been influenced by verbal descriptions of the contingencies generated under the negative slope conditions and occasioned in the yoked LVI condition.
In sum, the results of Experiment 2 are inconclusive with respect to the question of whether human behavior is sensitive to the presence or absence of the subtraction contingency. At least some of the problems may have been circumvented if the interpoint intervals were yoked across components of a multiple schedule rather than
across experimental conditions. The addition of discriminative stimuli for each of the different conditions may have helped foster sensitivity to the differences between the yoked and the negative slope conditions, and minimized the persistence of preestablished response patterns.
The overall pattern of results from both experiments suggests that human behavior is sensitive to the contingencies of negative-slope schedules. This was evidenced primarily by the decreases in overall rates of responding across negative-slope conditions in Experiment I (See Figure 3). In 22 of 23 negative-slope conditions across subjects, the steady-state response rates were lower than those of the preceding condition where the subtraction contingency was less stringent, indicating sensitivity to changes in the contingencies. These findings were partially replicated in Experiment 2 in that overall response rates in the negative slope conditions were lower than those of the initial VI conditions. The development of similar within-session response patterns across several subjects also indicates sensitivity to the negative-slope contingencies. Implications for Maximization Accounts
Sensitivity to the inverse relationship between overall response rate and overall
reinforcement rate was limited. Complete sensitivity to this relationship would have been evidenced by overall response rates that were consistently around 2 responses per minute. At this rate, all available points would have been collected in a timely fashion and the impact of the subtraction ratio on overall rate of point delivery would have been minimized. The results of Experiments I and 2 are therefore not in accord with
interpretations based strictly upon reinforcement maximization. In Figures 3 and 11, the data from the negative slope conditions do not lie near the apexes of the molar feedback functions for the respective schedules. By failing to respond at or around 2 responses per minute, the subjects failed to maximize overall rate of point delivery or to minimize the number of responses per point.
The human subjects in the present experiments fared better with respect to overall reinforcement density maximization than the pigeons in Vaughan and Miller's (1984) experiment, however. Across subjects, the response rates tended to fall closer to the apexes of the feedback function than those of the pigeons. Moreover, reinforcement rates occasionally increased for humans as the subtraction ratios became more stringent. whereas the pigeons tended to perform poorly at the lower subtraction ratios.
These results complement the results of experiments examining performance on
concurrent response-based and time-based schedules (Silberberg, Thomas, & Berendzen, 1991; Savastano & Fantino; 1994). On these procedures, reinforcement rate is maximized by responding mostly on the response-based alternative; the time-based schedule should be sampled only briefly and occasionally in order to collect reinforcers as they are set up. Both pigeons (Heyman & Herrnstein, 1986) and humans (Silberberg et al., 1991; Savastano & Fantino, 1994) allocate too much time to the time-based alternative and overall reinforcement rate is not maximized. Humans, however, tend to show a greater bias for the ratio alternative than do pigeons.
The present results are at odds with those of prior experiments examining "selfcontrol" (Logue et al., 1986) and choice in situations of diminishing returns (Hackenberg
& Axtell, 1992; Jacobs & Hackenberg, 1996; Wanchisen, et al., 1992) in humans, which are generally consistent with accounts based upon the maximization of overall reinforcement density. Indeed, deviations from maximization have become the exception rather than the rule in laboratory experiments with adult human subjects. It may be desirable to search these few exceptional cases for common features. This task is complicated, however, by significant methodological differences across experiments. In an experiment by Wasserman and Neunaber (1986), for example, pressing a telegraph key was maintained in student volunteers by contiguous point presentations even when pressing also decreased the overall rate of point delivery. These results suggest that human behavior can under certain circumstances be maintained by short-term consequences at the expense of long-term outcomes. It is important to note, however, that exposure to the contingencies was brief (1 session) and the points were not exchangeable for other reinforcers (i.e., they were merely "instructed" reinforcers). The effects can therefore be questioned on the grounds that the subjects did not have sufficient exposure to the contingencies to contact the long-term consequences, and that the point presentations were weak reinforcers (if indeed they were reinforcers at all). Implications for Matching, Accounts
Matching-based interpretations do not account well for the steady-state data either. As discussed above, the steady-state data from Experiment I portrayed in Figure 4 do not conform to Equation 2. This is in contrast with the results of several other studies examining human performance on VI schedules of reinforcement presented either alone or concurrently (Baum, 1975; Bradshaw, Szabadi, & Bevan, 1976, 1977, 1979;
McDowell & Wood, 1984; Shroeder & Holland, 1969: but see, Home & Lowe, 1993; Schmitt, 1974).
In two studies by Bradshaw and colleagues (1977, 1979) the effects of punishment on human variable-interval performance was examined. These studies warrant mention because, as in the present study, an inverse relationship between overall response rate and overall reinforcement rate was established. In these studies, button pressing was maintained in humans by points exchangeable for money. Point deliveries were arranged by a multiple schedule comprised of five different VI component schedules. Point delivery was signaled by a 100 ms illumination of a green light and the addition of one point to a counter. Under some conditions, a variable-ratio 34 schedule of point loss was in effect conjointly with the VI schedules. Every 34th response on average produced point loss signaled by a 100 ms illumination of a red lamp and the subtraction of one point from the counter.
Although there are many differences between this procedure and that of the negative slope conditions of the present study, both established an inverse relationship between overall response rate and overall reinforcement rate. The net result was the same: higher rates of responding resulted in lower rates of reinforcement overall. The major difference between the two procedures lies in the discriminability of point loss. In the Bradshaw et al. studies (1977, 1979), point loss was signaled, whereas the point omissions in the present study were unsignaled.
As in the negative-slope conditions of the present study, response rates in the
punishment conditions of the Bradshaw et al. studies (1977, 1979) were lower than
baseline rates. Unlike the results of Experiment 1, however, response rates in the punishment conditions varied as a function of reinforcement rate in a manner consistent with Equation 1. Relative to the data obtained in the unpunished conditions, the presence of the punishment contingency decreased L, the maximum rate of responding, and increased 10, the rate of reinforcement from unprogrammed sources. It would have been interesting to see if the VR schedule of point loss controlled within-session patterns of responding in a manner similar to the subtraction contingency of the negative slope conditions of the present experiment. Unfortunately, cumulative response records were not published.
Although the results of these and other studies by Bradshaw and colleagues lend support to the applicability of matching-based interpretations to human behavior, one aspect of the procedure complicates interpretation of the findings. As Home and Lowe (1993) pointed out, the discriminative stimuli that accompanied the different component schedules may have been functioning as implicit instructions, indicating differences in the schedule values. The stimuli consisted of a series of lights across the top of the apparatus. The left to right positioning of the lamps covaried with the VI schedule values; the light correlated with the richest schedule was on the extreme right, that correlated with the leanest schedule on the extreme left, with a graded series of schedule values in between.
Horne and Lowe (1993) performed a systematic replication of the Bradshaw et al.
(1976) study and produced evidence suggesting that the results could be explained on the basis of implicit ordinal stimuli. In a series of experiments, they found that matching was
frequently obtained when implicit or explicit ordinal cues were used as discriminative stimuli, but significant deviations from matching occurred when different geometrical shapes were projected onto a response panel as discriminative stimuli. Such results suggest that the performances in the Bradshaw et al. studies may have been verbally mediated through implicit instructions.
The Role of Verbal Mediation
Regarding verbal mediation, a possible explanation for the disparity between the results of the present experiment and those of the "self-control" procedures is that responding under the present circumstances was less influenced by verbal behavior. This research was undertaken, in part, to examine human behavior in an environment wherein the relationship between responding and point delivery was difficult to describe. It was hypothesized that the use of subtle contingencies of reinforcement would minimize the influence of verbal behavior on nonverbal performance. Although one subject (Subject 911, Exp. 2) offered a general description the relationship between rate of point delivery and rate of responding, none of the subjects were able to specify a maximization strategy as the subjects in the "self-control" experiments could (Logue et al., 1986). The periodicity of point delivery arranged by the present procedures may have been sufficiently subtle to preclude accurate descriptions, thereby eluding the influence of verbal mediation to a significant degree.
Because the present study was not designed to control verbal behavior nor to measure its relation to in nonverbal performance, it is not possible to rule out verbal mediation. To do so would require recording ongoing verbal behavior and its relation to the
nonverbal performance it accompanies. To avoid creating descriptions of the contingencies that might not otherwise have occurred, however, verbal reports were limited to a postexperimental questionnaire. Given the preliminary nature of the present research, it seemed prudent to examine contingency sensitivity with minimal verbal involvement before approaching the more complex interactions between verbal and nonverbal performance. A more precise characterization of the relationship between verbal behavior and performance under negative-slope schedules awaits further research. The Structure of Negative-Slope Schedule Performance
Although the present results are not well accounted for by interpretations based upon either the matching law or maximization, it is possible that order exists at a more local time scale. Reed and Schachtman (1991), for example, suggested that the difference between response rates on negative-slope schedules and yoked LVI schedules may be due to the differential reinforcement of relatively longer IRTs under the negative slope contingencies. In their experiment, interreinforcement intervals were yoked across negative slope and LVI components of a multiple schedule. They also required that the IRT preceding reinforcement in the yoked component fall within I s of that obtained in the negative slope component. Overall response rates were more similar across the two components when the IRT requirement was present than when it was absent. While this result lends indirect support to the hypothesis that negative slope schedules differentially reinforce longer IRTs, the results are far from unequivocal.
More convincing evidence would have been provided by examining differences in the IRT distributions in the two conditions. Unfortunately, none of the prior studies of
negative slope schedule performance provide data of this sort. In the present study, the IRTs preceding point deliveries were collected across all conditions for 6 of 7 subjects. The distribution of IRTs preceding point delivery became bimodal and the range increased under the negative slope conditions (See Figures 8, 9, & 14) in 4 of these 6 subjects. These data are consistent with Reed and Schachtman's (1991) hypothesis that relatively longer IRTs are differentially reinforced under negative slope schedules.
The bimodal distributions obtained in the present study reveal something that Reed and Schachtman's analysis could not. Merely stating that longer IRTs are reinforced under the negative slope schedules does not address the manner in which the distribution of IRTs changes. Such a statement would be as applicable to a shift in a unimodal distribution as it is to the present case. The fact that bimodal distributions of IRTs preceding point delivery occurred for several subjects suggests something about the structure of negative-slope schedule performance in general. Such performance appears to be characterized by alternating periods of pausing and high-rate responding. Verifying this relationship quantitatively would have required the collection of every IRT and the examination of shifts in the distributions across conditions. Unfortunately, this was not feasible in the present study. Short of recording every IRT, one can still gain some important insights regarding the moment-to-moment characteristics of negative slope schedule performance by examining cumulative response records.
Although cumulative records are not published as frequently as they once were (Skinner, 1976), they are still useful for assessing schedule control by portraying schedule-typical moment-to-moment fluctuations in responding (Ferster & Skinner,
1957; Zeiler, 1984). In all studies employing negative-slope schedules to date (Ettinger, et al., 1987; Reed & Schachtman, 1989; 1991; Vaughan & Miller, 1984) however, there has been only one cumulative record published (Ettinger, et al., 1987) and it is not particularly useful. In that study, lever pressing was maintained in rats by an LVI 90 s schedule of food presentation, with food deliveries conjointly omitted according to an FR 20 schedule. A composite cumulative record of the first 20 responses following food presentation averaged across the entire session was published. As a result of the limited range of data presented and the curve smoothing effects of the transform, the record is of little use in discerning the characteristic moment-to-moment fluctuations in responding generated by the schedule. The absence of cumulative records of negative-slope performance in the literature makes it difficult to determine if adequate schedule control had been achieved.
In the present study, cumulative records were obtained for all subjects under all
conditions. Across the two experiments, the negative slope schedules generated a "pause and run" pattern of responding in 4 of 7 subjects. The onset of responding was usually met with a series of point deliveries until the "store" was depleted. Responding often continued after the "store" was depleted and additional points were occasionally delivered. This pattern differs considerably from that typical of VI performance, characterized by responding at a constant rate with little pausing. For the four subjects (Subjects I 11, 888, 211, and 52 1) showing this pattern, typical VI patterning was established in the initial VI condition, was maintained throughout the LVI condition, then transitioned to the "pause and run" pattern across negative-slope conditions. The
development of this pattern provides further evidence that the subjects' behavior was sensitive to the subtraction contingency. If responding was simply tracking changes in overall rate of point delivery, there is no reason to suppose that response patterns would differ across the VI and negative-slope conditions. Indeed, one might expect differences to be seen primarily in the overall rates, not in the local patterning.
Of the 3 subjects (Subjects 999, 711, and 911) that did not show the "pause-run"
pattern robustly, all responded at much lower rates in the baseline VI and LVI conditions than the other subjects, with two (Subjects 999 and 911) showing sensitivity to the VlLVI difference. Moreover, these subjects generally received less exposure overall to the negative slope schedules than some of the other subjects. It is possible that these subjects did not receive sufficient exposure to the negative slope schedules for the "pause-run" pattern to develop.
It is also possible that the "pause-run" pattern of responding depends on the baseline response rate. The response patterns of humans on fixed interval schedules, for example, generally fall into one of two categories: low rate, characterized by long pauses with a few responses at the end of the interval, or high rate, characterized by infrequent pausing and virtual insensitivity to schedule changes (Laties & Weiss, 1963; Lippman & Meyer, 1967). Weiner (1969) has demonstrated that a prior experimental history can influence the pattern of FI responding. Subjects initially exposed to an IRT > t s schedule tended to respond at a low rate when subsequently exposed to a series of FI schedules, whereas those initially exposed to a fixed-ratio schedule tended to respond at a high constant rate under the Fl schedules. It is possible that the subjects' preexperimental histories and/or
their low-rate histories on the baseline VI and LVI conditions modulated sensitivity to the negative-slope conditions of the present experiment.
Cumulative response records may be a useful tool for the identification of response units created and maintained under negative slope schedules. The "pause and run" response pattern may be one type of unit characteristic of negative-slope schedule performance. Although the reliability of this pattern across procedures and species needs to be established through replication, examination of these regularities in responding may prove useful in the identification of appropriate behavioral units and the development of quantitative descriptions that improve prediction and control. In future research, every IRT should be recorded to examine cyclical fluctuations in response rate as they are related to cyclical variations in the temporal distribution of reinforcement. By collecting every IRT, behavioral units across different time scales can be examined. The "pause and run" pattern itself, for example, may be the aggregate effect of differential reinforcement acting on different classes of IRTs, or on some local property of responding. Mathematical models formulated to account for the local dynamics of behavior (e.g., Killeen, 1994) may one day be brought to bear on negative slope performance as well.
SUMMARY AND CONCLUSIONS
Previous research with human subjects has implicated verbal behavior as a possible
determinant of between-species differences in sensitivity to temporally remote outcomes. The results of the present study tentatively support this position and suggest one possible method for minimizing the influence of verbal behavior on nonverbal performance. Human behavior was examined under contingencies that did not give rise to accurate, quantitative descriptions of the relationship between responding and point delivery and results more similar to those of nonhumans were obtained. The results suggest that human behavior may be more likely to be controlled by immediate outcomes when it is difficult to describe the relationship between responding and consequences. Due to limitations in the resolution of data collection, however, this study should be regarded as preliminary. Because assessment of verbal behavior was limited in the present study, additional research is needed to evaluate the role of verbal behavior in establishing and maintaining sensitivity to temporally remote outcomes. Additional research is also necessary to provide a precise quantitative account of negative-slope schedule performance.
Ainslie, G. W. (1974). Impulse control in pigeons. Journal of the Experimental Analysis
of Behavior, 21, 485-489.
Baum, W. M. (1973). The correlation-based law of effect Journal of the Experimental
Analysis of Behavior, 20, 137-153.
Baum, W. M. (1975). Time allocation in human vigilance. Journal of the Experimental
Analysis of Behavior, 23, 45-53.
Bradshaw, C. M., Szabadi, E, & Bevan, P. (1976). Behavior of humans in variableinterval schedules of reinforcement. Journal of the Experimental Analysis of
Behavior, 27, 275-279.
Bradshaw, C. M., Szabadi, E, & Bevan, P. (1977). Effect of punishment on human
variable-interval performance. Journal of the Experimental Analysis of Behavior,
Bradshaw, C. M., Szabadi, E, & Bevan, P. (1979). The effect of punishment on freeoperant choice in humans. Journal of the Experimental Analysis of Behavior, 31,
Catania, A. C., Mathews, B. A., & Shimoff, E. (1982). Instructed versus shaped human
verbal behavior: Interactions with nonverbal responding. Journal of the
Experimental Analysis of Behavior, 38, 233-248.
Catania, A. C., & Reynolds, G. S. (1968). A quantitative analysis of the responding
maintained by interval schedules of reinforcement. Journal of the Experimental
Analysis of Behavior, 11, 327-383.
Cerutti, D. T. (1989). Discriminative theory of rule-governed behavior. Journal of the
Experimental Analysis of Behavior, 51, 259-276.
Ettinger, R. H., Ried, A. K., & Staddon, J. E. R. (1987). Sensitivity to Molar Feedback
Functions: A Test of Molar Optimality Theory. Journal of Experimental
Psychology: Animal Behavior Processes, 13, 366-375.
Ferster, C. B., & Skinner, B.F. (1957). Schedules of reinforcement. New York:
Fleshier M., & Hoffman, H. S., (1962). A progression for generating variable-interval
schedules. Journal of the Experimental Analysis of Behavior, 5, 529-530.
Hackenberg, T. D., & Axtell, S. A. M. (1993). Humans' choices in situations of timebased diminishing returns. Journal of the Experimental Analysis of Behavior.
Hackenberg, T. D., & Hineline, P. N. (1992). Choice in situations of diminishing
returns: Immediate versus delayed consequences of action. Journal of the
Experimental Analysis of Behavior, 57, 67-80.
Herrnstein, R. J. (1970). On the law of effect. Journal of the Experimental Analysis of
Behavior, 13, 243-266.
Heyman, G. M., & Hermstein, R. J., (1986). More concurrent interval-ratio schedules:
A replication and review. Journal of the Experimental Analysis of Behavior, 46.
Hineline, P. N., & Sodetz, F. J. (1987). Appetitive and aversive schedule preferences:
Schedule transitions as intervening events. In MA L. Commons, J. E. Mazur, J. A.
Nevin, & H. Rachlin (Eds.), Quantitative analyses of behavior: Vol. 5. The effect of delay and intervening events on reinforcement value (pp. 141-157). Hillsdale,
Hodos, W., & Trumbule, G. H. (1967). Strategies of schedule preference in
chimpanzees. Journal of the Experimental Analysis of Behavior, 10. 503-514.
Horne, P. J., & Lowe, C. F. (1993). Determinants of human performance on concurrent
schedules. Journal of the Experimental Analysis of Behavior, 59, 29-60.
Home, P. J., & Lowe, C. F. (1996). On the origins of naming and other symbolic
behavior. Journal of the Experimental Analysis of Behavior, 65, 185-241.
Hyten, C., Madden, G. J., and Field, D. P. (1994). Exchange delays and impulsive
choice in adult humans. Journal of the Experimental Analysis of Behavior, 62,
Jackson, K., & Hackenberg, T. D. (1996). Token reinforcement, choice, and self-control
in pigeons. Journal of the Experimental Analysis of Behavior, 66, 29-49.
Jacobs, E. A., & Hackenberg, T. D., (1996). Humans' choices in situations of
diminishing returns: Effects of fixed-interval duration and progressive-interval
step size. Journal of the Experimental Analysis of Behavior, 65, 5-19.
Logue, A. W. (1988). Research on self-control: An integrating framework. Behavioral
& Brain Sciences, 11, 665-709.
Logue, A.W., & King, G.R. (1991). Self-control and impulsiveness in adult humans
when food is the reinforcer. Appetite, 17, 105-120.
Logue, A.W., Pena-Correal, T.E., Rodriguez, M.L., & Kabela, E. (1986). Self-control in
adult humans: Variation in positive reinforcer amount and delay. Journal of the
Experimental Analysis of Behavior, 46, 159-173.
Killeen P. (1994). Mathematical principles of reinforcement. Behavioral and Brain
Science, 17, 105-172.
Mazur, J. E., & Vaughan, W. Jr. (1987). Molar optimization versus delayed
reinforcement as explanations of choice between fixed-ratio and progressive-ratio
schedules. Journal of the Experimental Analysis of Behavior, 48, 251-261.
McDowell, J.J., & Wood, H.M. (1984). Confirmation of linear systems theory
prediction: Changes in Herrnstein's k as a function of changes in reinforcer magnitude. Journal of the Experimental Analysis of Behavior, 41, 183-192.
Navarick, D. J. (1982). Negative reinforcement and choice in humans. Learning and
Motivation, 13, 361-377.
Prelec, D. (1982). Matching, maximizing, and the hyperbolic reinforcement feedback
function. Psychological Review, 89, 189 230.
Rachlin, H., & Green, L. (1972). Commitment, choice, and self-control. Journal of the
Experimental Analysis of Behavior, 17, 15-22.
Reed, P., & Schachtman, T. R. (1991). Instrumental performance on negative schedules.
The Quarterly Journal of Experimental Psychology, 43b, 177-197.
Reed, P., & Schachtman, T. R. (1989). Instrumental responding by rats on free operant
contingencies with components that schedule response-dependent reinforcer
omission. Animal Learning and Behavior, 17, 328-338.
Rosenfarb, I. S., Newland, M. C., Brannon, S. E., & Howey, D. S. (1992). Effects of
self-generated rules on the development of schedule-controlled behavior. Journal
of the Experimental Analysis of Behavior, 58, 107-121.
Savastano, H., & Fantino, E. F. (1994). Human choice in concurrent ratio-interval
schedules of reinforcement. Journal of the Experimental Analysis of Behavior,
Schmitt, D.R. (1974). Effects of reinforcement rate and reinforcer magnitude on choice
behavior of humans. Journal of the Experimental Analysis of Behavior, 21, 409419.
Schroeder, S.R., & Holland, J.G., (1969). Reinforcement of eye movement with
concurrent schedules. Journal of the Experimental Analysis of Behavior, 12, 897903.
Shimoff, E., Matthews, B.A., & Catania, A.C. (1986). Human operant performance:
Sensitivity and pseudosensitivity to contingencies. Journal of the Experimental
Analysis of Behavior, 46, 149-157.
Silberberg, A., Thomas, J. R., & Berendzen, N. (1991). Human choice on concurrent
variable-interval, variable-ratio schedules. Journal of the Experimental Analysis
of Behavior, 56, 575-584.
Skinner, B. F. (1957). Verbal behavior, New York: Appleton-Century-Crofts.
Skinner, B. F. (1969). An operant analysis of problem solving. In K. MacCorquodale,
G. Lindzey, & K. E. Clark (Eds.), Cumulative record (pp. 295-313).
Skinner, B. F. (1976). Farewell, my lovely. Journal of the Experimental Analysis of
Behavior, 25, 218.
Solnick, J. V., Kannenberg, C.H., Eckerman, D. A., & Waller, M. B. (1980). An
experimental analysis of impulse control in humans, Learning and Motivation, 11,
Sonuga-Barke, E.J.S., Lea, S.E.G., & Webley, P.(1989). The development of adaptive
choice in a self-control paradigm. Journal of the Experimental Analysis of
Behavior, 51, 77-85.
Vaughan, W., Jr. (1982). Choice and the Rescorla-Wagner model. In M. L. Commons,
R. J. Herrnstein, & H. Rachlin (Eds.) Quantitative analyses of behavior: Vol. 2.
Matching and maximizing accounts (pp. 263 -279). Cambridge, MA: Ballinger.
Vaughan, W., Jr. & Miller, H. L. (1984). Optimization versus response strength
accounts of behavior. Journal of the Experimental Analysis of Behavior. 42, 337348.
Wanchisen, B. A., Tatham, T. A., & Hineline, P. N. (1988). Pigeons'choices in
situations of diminishing returns: Fixed-versus progressive-ratio schedules.
Journal of the Experimental Analysis of Behavior, 50, 375-394.
Wanchisen, B. A., Tatham, T. A., & Hineline, P. N. (1992). Human choice in
counterintuitivev" situations: Fixed- versus progressive schedules. Journal of the
Experimental Analysis of Behavior, 58, 67-85.
Wasserman, E. A., & Neunaber, D. J., (1986). College students' responding to and
rating of contingency relations: The role of temporal contiguity. Journal of the
Experimental Analysis of Behavior, 46, 15-35.
Weiner, H. (1969). Controlling human fixed-interval performance. Journal of the
Experimental Analysis of Behavior, 12, 349-373.
Weiner, H. (1970). Human behavioral persistence. The Psychological Record, 20, 445456.
Weiner, H. (1983). Some thoughts on discrepant human-animal performances under
schedules of reinforcement. The Psychological Record, 33, 52 1-532.
Zeiler, M.D. (1984). Schedules of reinforcement: The sleeping giant. Journal of the
Experimental Analysis of Behavior, 42, 485-493.
Eric A. Jacobs was born in South Amboy, New Jersey, on May 17, 1967. When he
was 12 years old, the Jacobs family moved to Florida. After graduating from Palm Beach Gardens Community High School in 1985, Eric attended the University of Florida where he studied psychology. In 1990, he received a Bachelor of Science degree and began graduate studies under the direction of Timothy D. Hackenberg, Ph.D. In 1994, he was awarded the degree of Master of Science. Eric currently resides with his wife, Angela Rabb Jacobs, in Burlington, Vermont, where he serves as an Assistant Project Director for the Substance Abuse Treatment Center at the University of Vermont. Eric longs for the camaraderie of colleagues at the University of Florida, for the simplicity of microswitches, for the clatter of cumulative response recorders, and for the elegance of within-subject designs.
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E6OMREWUQ_T088Z1 INGEST_TIME 2014-10-06T21:44:25Z PACKAGE AA00025776_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
+80$1 3(5)250$1&( 21 1(*$7,9(6/23( 6&+('8/(6 2) 5(,1)25&(0(17 $ 7(67 2) 5(,1)25&(0(17 0$;,0,=$7,21 %\ (5,& $ -$&2%6 $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$
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
7$%/( 2) &217(176 SDJH $&.12:/('*0(176 $%675$&7 Y ,1752'8&7,21 (;3(5,0(17 0HWKRG 6XEMHFWV $SSDUDWXV 3URFHGXUH ([SHULPHQWDO &RQGLWLRQV 3URFHGXUDO ,UUHJXODULWLHV 5HVXOWV 0RODU )HHGEDFN )XQFWLRQV 6HVVLRQE\6HVVLRQ 3DWWHUQV :LWKLQ6HVVLRQ 3DWWHUQV ,57V 3UHFHGLQJ 3RLQW 'HOLYHU\ 'LVFXVVLRQ (;3(5,0(17 0HWKRG 6XEMHFWV $SSDUDWXV 3URFHGXUH ([SHULPHQWDO &RQGLWLRQV 3URFHGXUDO ,UUHJXODULWLHV 5HVXOWV 0RODU )HHGEDFN )XQFWLRQV 6HVVLRQE\6HVVLRQ 3DWWHUQV :LWKLQ6HVVLRQ 3DWWHUQV ,57V 3UHFHGLQJ 3RLQW 'HOLYHU\ LLL
'LVFXVVLRQ *(1(5$/ ',6&866,21 ,PSOLFDWLRQV IRU 0D[LPL]DWLRQ $FFRXQWV ,PSOLFDWLRQV IRU 0DWFKLQJ $FFRXQWV 7KH 5ROH RI 9HUEDO 0HGLDWLRQ 7KH 6WUXFWXUH RI 1HJDWLYH6ORSH 6FKHGXOH 3HUIRUPDQFH 6800$5< $1' &21&/86,216 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ ,9
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
,1752'8&7,21 7KH GHJUHH WR ZKLFK EHKDYLRU LV VHQVLWLYH WR VKRUWWHUP YHUVXV ORQJHUWHUP FRQVHTXHQFHV LV DQ DUHD RI UHVHDUFK ZLWK LPSRUWDQW LPSOLFDWLRQV IRU EDVLF DQG DSSOLHG UHVHDUFKHUV )RU WKH EDVLF UHVHDUFKHU VXFK VWXGLHV FDQ SURYLGH FULWLFDO LQIRUPDWLRQ DERXW WKH WLPH VFDOH RYHU ZKLFK FRQVHTXHQFHV DUH IXQFWLRQDOO\ UHODWHG WR EHKDYLRU ZKLFK LV LPSRUWDQW LQ XQGHUVWDQGLQJ KRZ UHLQIRUFHPHQW FUHDWHV DQG PDLQWDLQV EHKDYLRUDO XQLWV )RU WKH DSSOLHG UHVHDUFKHU VXFK VWXGLHV PD\ EH XVHIXO LQ XQGHUVWDQGLQJ EHKDYLRUDO SUREOHPV WKDW VWHP IURP H[FHVVLYH FRQWURO E\ SUR[LPDO FRQVHTXHQFHV )RU H[DPSOH SHRSOH DEXVH GUXJV JDPEOH H[FHVVLYHO\ LQFXU ODUJH DPRXQWV RI GHEW DQG HDW SRRUO\ ZLWKRXW VHHPLQJ UHJDUG IRU WKH WHPSRUDOO\ GHIHUUHG FRQVHTXHQFHV RI WKHVH DFWLYLWLHV 7KH DELOLW\ WR GHPRQVWUDWH DQG PDQLSXODWH VLPLODU LPSXOVLYH EHKDYLRU SDWWHUQV LQ D FRQWUROOHG ODERUDWRU\ VHWWLQJ PD\ JUHDWO\ LQFUHDVH RXU XQGHUVWDQGLQJ RI WKH FRQGLWLRQV XQGHU ZKLFK WKH\ RFFXU ZKLFK PD\ LQ WXUQ PDNH LW SRVVLEOH WR FRUUHFW RU SUHYHQW WKH RFFXUUHQFH RI VXFK EHKDYLRU RXWVLGH RI WKH ODERUDWRU\ 7KH PRVW ZLGHO\ XVHG SURFHGXUHV WR DVVHVV FRQWURO E\ LPPHGLDWH YHUVXV ORQJWHUP FRQVHTXHQFHV LQYROYH PXWXDOO\ H[FOXVLYH FKRLFHV EHWZHHQ D VPDOO LPPHGLDWH UHLQIRUFHU DQG D ODUJHU GHOD\HG UHLQIRUFHU $LQVOLH /RJXH 5DFKOLQ t *UHHQ f :KHQ EHKDYLRU LV JRYHUQHG E\ WKH VPDOOHU LPPHGLDWH UHLQIRUFHU WKH SHUIRUPDQFH LV VDLG WR EH LPSXOVLYH RU WR GHPRQVWUDWH LPSXOVLYLW\f &RQYHUVHO\ ZKHQ EHKDYLRU LV
JRYHUQHG E\ WKH ODUJHU GHOD\HG UHLQIRUFHU WKH SHUIRUPDQFH LV VDLG WR EH VHOIFRQWUROOHG RU WR GHPRQVWUDWH VHOIFRQWURO 2QH RI WKH DGYDQWDJHV RI VXFK SURFHGXUHV LV WKDW WKH\ KDYH EHHQ XVHG SURILWDEO\ ZLWK KXPDQ DQG QRQKXPDQ VXEMHFWV DOLNH ,Q D FRPSDUDWLYH UHYLHZ RI WKH VHOIFRQWURO OLWHUDWXUH /RJXH f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f ZLWK FKLPSDQ]HHV WKH SURFHGXUH LQYROYHV UHSHDWHG FKRLFHV EHWZHHQ D IL[HG VFKHGXOH WKH UHTXLUHPHQWV RI ZKLFK UHPDLQ FRQVWDQW ZLWKLQ D VHVVLRQ DQG D SURJUHVVLYH VFKHGXOH WKH UHTXLUHPHQWV RI ZKLFK DUH LQLWLDOO\ ORZHU WKDQ WKH IL[HG EXW JUDGXDOO\ EHFRPH PRUH VWULQJHQW ZLWK HDFK UHLQIRUFHPHQW GHOLYHUHG E\ WKDW VFKHGXOH
7KH GHSHQGHQW PHDVXUH RI PRVW LQWHUHVW XQGHU WKHVH SURFHGXUHV LV WKH SDWWHUQ RI VZLWFKLQJ IURP WKH SURJUHVVLYH VFKHGXOH WR WKH IL[HG 8QGHU VRPH FRQGLWLRQV VKRUWWHUP FRQVHTXHQFHV WKRVH EDVHG VROHO\ XSRQ PLQLPL]LQJ WKH GHOD\ WR WKH XSFRPLQJ UHLQIRUFHUf DQG ORQJWHUP FRQVHTXHQFHV WKRVH EDVHG XSRQ PD[LPL]LQJ RYHUDOO UHLQIRUFHPHQW UDWHf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t 6RGHW] f SLJHRQV +DFNHQEHUJ t +LQHOLQH 0D]XU t 9DXJKDQ :DQFKLVHQ 7DWKDP t +LQHOLQH f DQG KXPDQV +DFNHQEHUJ t $[WHOO -DFREV t +DFNHQEHUJ :DQFKLVHQ 7DWKDP t +LQHOLQH f LQ DGGLWLRQ WR WKH RULJLQDO VWXG\ ZLWK FKLPSDQ]HHV +RGRV t 7UXPEXOH f $V ZLWK VHOIFRQWURO SURFHGXUHV KRZHYHU KXPDQ SHUIRUPDQFHV RQ WKHVH SURFHGXUHV DUH JHQHUDOO\ EHWWHU GHVFULEHG E\ DFFRXQWV EDVHG XSRQ WKH PD[LPL]DWLRQ RI UHLQIRUFHPHQW GHQVLW\ ZKHUHDV WKH SHUIRUPDQFHV RI QRQKXPDQV DUH JHQHUDOO\ EHWWHU FKDUDFWHUL]HG E\ D GHOD\ EDVHG DFFRXQW +DFNHQEHUJ t $[WHOO 0D]XU t 9DXJKDQ f $OWKRXJK WKH UHVXOWV RI WKHVH FKRLFH H[SHULPHQWV KDYH UHYHDOHG WKDW QRQKXPDQ EHKDYLRU FDQ FRPH XQGHU FRQWURO RI WHPSRUDOO\ UHPRYHG FRQVHTXHQFHV WR VRPH H[WHQW WKH\ DOVR VXJJHVW WKDW
KXPDQ EHKDYLRU PD\ EH PRUH VHQVLWLYH WKDQ QRQKXPDQ EHKDYLRU WR ORQJWHUP RYHUDOO RXWFRPHV 7KH EHWZHHQVSHFLHV GLVSDULW\ LQ WKH UHVXOWV RI WKH VHOIFRQWURO H[SHULPHQWV GHVFULEHG DERYH PD\ UHIOHFW D IXQGDPHQWDO GLIIHUHQFH EHWZHHQ KXPDQV DQG RWKHU VSHFLHV LQ WKH VHQVLWLYLW\ WR GHIHUUHG FRQVHTXHQFHV ,W LV GLIILFXOW KRZHYHU WR HVWDEOLVK XQHTXLYRFDOO\ WKDW GLIIHUHQFHV LQ SHUIRUPDQFH DUH VROHO\ RI SK\ORJHQLF RULJLQ 7R GR VR RQH ZRXOG KDYH WR UXOH RXW DW OHDVW WZR RWKHU SRVVLELOLWLHV Df WKDW VSHFLHV GLIIHUHQFHV ZHUH QRW GXH WR GLIIHUHQFHV LQ SURFHGXUH RU Ef WKDW VSHFLHV GLIIHUHQFHV ZHUH QRW GXH WR GLIIHUHQFHV LQ SUHH[SHULPHQWDO KLVWRULHV 7KHVH ZLOO EH FRQVLGHUHG LQ WXUQ 2I WKH PDQ\ SURFHGXUDO GLIIHUHQFHV EHWZHHQ H[SHULPHQWV ZLWK KXPDQV DQG WKRVH ZLWK QRQKXPDQV GLIIHUHQFHV LQ WKH W\SHV RI UHLQIRUFHUV PD\ FRQWULEXWH WR EHWZHHQVSHFLHV SHUIRUPDQFH GLIIHUHQFHV 6WXGLHV ZLWK QRQKXPDQV W\SLFDOO\ XVH XQFRQGLWLRQHG UHLQIRUFHUV HJ JUDLQ SUHVHQWDWLRQ WR IRRG GHSULYHG SLJHRQVf ZKHUHDV H[SHULPHQWV ZLWK KXPDQV W\SLFDOO\ XVH WRNHQ UHLQIRUFHUV SRLQWV H[FKDQJHDEOH IRU PRQH\f ,Q VHOI FRQWURO H[SHULPHQWV IRU H[DPSOH KXPDQV PDNH UHSHDWHG FKRLFHV EHWZHHQ D VPDOO DPRXQW RI SRLQWV H[FKDQJHDEOH IRU PRQH\ GHOLYHUHG LPPHGLDWHO\ DQG D ODUJHU DPRXQW RI SRLQWV GHOLYHUHG DIWHU D EULHI GHOD\ 7KH GHOD\ WR H[FKDQJH WKH SRLQWV IRU PRQH\ LV XVXDOO\ IL[HG DQG UHPDLQV WKH VDPH LH DW VHVVLRQnV HQGf UHJDUGOHVV RI WKH FKRLFH SDWWHUQV +XPDQV PD\ FRQVLVWHQWO\ SUHIHU WKH GHIHUUHG ODUJHU DPRXQW RI SRLQWV EHFDXVH WKHUH LV QR LPPHGLDWH DGYDQWDJH WR REWDLQLQJ SRLQWV TXLFNO\ DV WKH\ DUH QRW H[FKDQJHG IRU FRQVXPDEOH UHLQIRUFHUV XQWLO ODWHU 7KLV K\SRWKHVLV LV VXSSRUWHG E\ WKH IDFW WKDW LPSXOVLYH SDWWHUQV LQ KXPDQV DUH PRUH OLNHO\ ZLWK UHLQIRUFHUV WKDW DUH PRUH LPPHGLDWHO\
HIIHFWLYH VXFK DV HVFDSH IURP ZKLWH QRLVH 1DYDULFN 6ROQLFN .DQQHQEHUJ (FNHUPDQ t :DOOHU f RU DFFHVV WR HGLEOH UHLQIRUFHUV /RJXH t .LQJ f 0RUHRYHU UHFHQW H[SHULPHQWV E\ +\WHQ 0DGGHQ DQG )LHOG f DQG E\ -DFNVRQ DQG +DFNHQEHUJ f KDYH GHPRQVWUDWHG WKDW VLPLODU SHUIRUPDQFHV LQ SHRSOH DQG SLJHRQV UHVSHFWLYHO\ FDQ EH DFKLHYHG E\ PDQLSXODWLQJ H[FKDQJH GHOD\V LQ VHOIFRQWURO SURFHGXUHV LQYROYLQJ WRNHQ UHLQIRUFHUV (YHQ ZKHQ SURFHGXUDO GLIIHUHQFHV DUH PLQLPL]HG WKH SRVVLELOLW\ UHPDLQV WKDW KXPDQ RSHUDQW SHUIRUPDQFH LQ WKH ODERUDWRU\ PD\ LQWHUDFW ZLWK D VXEMHFWnV SULRU YHUEDO KLVWRU\ &DWDQLD 0DWKHZV t 6KLPRII +RPH t /RZH 5RVHQIDUE 1HZODQG %UDQQRQ t +RZH\ f +XPDQV QRW RQO\ UHVSRQG WR FRQWLQJHQFLHV EXW GHVFULEH WKRVH FRQWLQJHQFLHV DQG KRZ WKH\ DUH UHODWHG WR EHKDYLRU 6NLQQHU f LQWURGXFHG D GLVWLQFWLRQ EHWZHHQ UXOHJRYHUQHG DQG FRQWLQJHQF\VKDSHG EHKDYLRU WR KHOS FKDUDFWHUL]H WKH GLIIHUHQW ZD\V LQ ZKLFK KXPDQ EHKDYLRU FRPHV XQGHU FRQWURO RI WKH HQYLURQPHQW &RQWLQJHQF\VKDSHG EHKDYLRU LV EHKDYLRU WKDW LV JRYHUQHG E\ GLUHFW H[SHULHQFH ZLWK FRQWLQJHQFLHV 5XOHJRYHUQHG EHKDYLRU RQ WKH RWKHU KDQG LV EHKDYLRU WKDW LV XQGHU WKH FRQWURO RI FRPSOH[ DQWHFHGHQW VWLPXOL HVWDEOLVKHG WKURXJK D KLVWRU\ RI VRFLDOO\ PHGLDWHG UHLQIRUFHPHQW &HUXWWL +RPH t /RZH 6NLQQHU f 1RUPDO DGXOW KXPDQV HQWHU WKH ODERUDWRU\ ZLWK WKH DELOLW\ QRW RQO\ WR IROORZ UXOHV EXW DOVR WR GHULYH UXOHV GHVFULELQJ WKH UHODWLRQVKLS EHWZHHQ EHKDYLRU DQG FRQVHTXHQFHV 9HUEDO KXPDQV PD\ DXJPHQW WKH HIIHFWV RI FRQWLQJHQFLHV ZLWK YHUEDO GHVFULSWLRQV RI WKRVH FRQWLQJHQFLHV DV D UHVXOW WKHLU SHUIRUPDQFHV PD\ GLIIHU IURP WKRVH RI QRQYHUEDO VXEMHFWV
/LNH QRQYHUEDO EHKDYLRU UXOHIROORZLQJ DQG UXOHJHQHUDWLQJ DUH FODVVHV RI EHKDYLRU HVWDEOLVKHG DQG PDLQWDLQHG E\ FRQVHTXHQFHV )RU H[DPSOH VXFK DELOLWLHV PD\ EH KHOSIXO IRU HVWDEOLVKLQJ UHVSRQGLQJ LQ DQRWKHU PHPEHU RI WKH YHUEDO FRPPXQLW\ LQ RFFDVLRQLQJ D FRPSOLFDWHG SHUIRUPDQFH WKDW PD\ KDYH EHHQ RWKHUZLVH IRUJRWWHQ RU LQ JHQHUDOL]LQJ EHKDYLRU WR FRQWLQJHQFLHV WKDW EHDU VRPH VLPLODULW\ WR WKRVH SUHYLRXVO\ H[SHULHQFHG 6NLQQHU f VXJJHVWHG WKDW DQRWKHU SRVVLEOH EHQHILW RI UXOHJRYHUQHG EHKDYLRU LV WR EULQJ EHKDYLRU LQWR FRQWDFW ZLWK WHPSRUDOO\ GHIHUUHG RXWFRPHV $FFRUGLQJ WR KLV YLHZ UXOHV PD\ EH KHOSIXO LQ VXVWDLQLQJ SHUIRUPDQFHV WKDW ZRXOG QRUPDOO\ EH FXUWDLOHG E\ PRUH SUR[LPDO FRQVHTXHQFHV 7KLV PD\ DFFRXQW IRU WKH EHWZHHQVSHFLHV GLIIHUHQFHV LQ SHUIRUPDQFH REVHUYHG LQ H[SHULPHQWV WKDW SODFH ORQJWHUP DQG VKRUWWHUP FRQVHTXHQFHV LQ RSSRVLWLRQ 7KHUH LV VRPH HYLGHQFH LQ WKH OLWHUDWXUH WR VXSSRUW WKLV SRVLWLRQ $GXOW KXPDQV LQ D VHOIFRQWURO H[SHULPHQW IRU H[DPSOH UHSRUWHG XVLQJ PD[LPL]DWLRQ VWUDWHJLHV WKDW ZHUH GHULYHG E\ FRXQWLQJ GXULQJ WKH LQWHUYDOV /RJXH 3HD&RUUHDO 5RGULJXH] t .DEHOD f 7KDW LV WKH VXEMHFWV ZHUH DEOH WR GHVFULEH WKH RYHUDOO UHODWLRQVKLS EHWZHHQ WKHLU FKRLFHV DQG WKH DPRXQW RI UHLQIRUFHPHQW 6XFK GHVFULSWLRQV PD\ KDYH IXQFWLRQHG DV YHUEDO GLVFULPLQDWLYH VWLPXOL WKDW EURXJKW WKHLU EHKDYLRU LQWR FRQWDFW ZLWK WKH GHIHUUHG FRQVHTXHQFHV $GGLWLRQDO HYLGHQFH RI WKH UROH RI YHUEDO EHKDYLRU LQ HVWDEOLVKLQJ FRQWURO E\ GHIHUUHG FRQVHTXHQFHV FRPHV IURP D VWXG\ RI DJHUHODWHG GLIIHUHQFHV LQ VHOIFRQWURO E\ 6RQXJD %DUNH /HD DQG :HEOH\ f ,Q WKDW VWXG\ JLUOV DJHV DQG PDGH UHSHDWHG FKRLFHV EHWZHHQ D VPDOO UHLQIRUFHU RQH WRNHQf DQG D ODUJHU UHLQIRUFHU WZR WRNHQVf 7KH
GHOD\ WR WKH VPDOO UHLQIRUFHU ZDV KHOG FRQVWDQW DW V ZKHUHDV WKH GHOD\ WR WKH ODUJH UHLQIRUFHU ZDV YDULHG IURP V WR V DFURVV FRQGLWLRQV 2YHUDOO UHLQIRUFHPHQW PD[LPL]DWLRQ ZDV WKXV FRQWLQJHQW XSRQ SUHIHUULQJ WKH VPDOOHU PRUH LPPHGLDWH UHLQIRUFHU ZKHQ WKH GHOD\ WR WKH ODUJHU ZDV JUHDWHU WKDQ V 7KH UHVXOWV LQGLFDWHG GHYHORSPHQWDO GLIIHUHQFHV WKDW FRUUHVSRQGHG ZLWK WKH VXEMHFWVf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
f 7UDLQLQJ VRPHRQH WR GULYH D FDU ZLWK D PDQXDO WUDQVPLVVLRQ SURYLGHV DQ H[DPSOH RI WKH XVH RI UXOH JRYHUQDQFH WR EULQJ EHKDYLRU XQGHU WKH FRQWURO RI D SDUWLFXODU VHW RI FRQWLQJHQFLHV 7KH QRYLFH GULYHU LV RIWHQ JLYHQ LQVWUXFWLRQV RQ KRZ WR RSHUDWH WKH FOXWFK DQG WKH DFFHOHUDWRU VLPXOWDQHRXVO\ 7KH LQVWUXFWRU PD\ VD\ 6ORZO\ UHOHDVH WKH FOXWFK ZKLOH VLPXOWDQHRXVO\ GHSUHVVLQJ WKH DFFHOHUDWRU $OWKRXJK VXFK LQVWUXFWLRQ LV OLNHO\ WR EH RI VRPH DVVLVWDQFH LQ JHQHUDWLQJ EHKDYLRU VXIILFLHQW WR PRYH WKH FDU LW LV KLJKO\ XQOLNHO\ WKDW D ZHOO H[HFXWHG SHUIRUPDQFH ZLOO RFFXU RQ WKH ILUVW DWWHPSW $ IOXLG GHSDUWXUH ZLOO RFFXU RQO\ DIWHU VXIILFLHQW H[SHULHQFH ZLWK WKH FRQWLQJHQFLHV LPSRVHG E\ WKH FDU LWVHOI 7KH LQVWUXFWRU FDQQRW RIIHU D SUHFLVH GHVFULSWLRQ RI DSSURSULDWH SHUIRUPDQFH EDVHG XSRQ KLV RU KHU HYHU\GD\ LQWHUDFWLRQ ZLWK WKH GHYLFH 6ORZO\ LV D UDWKHU YDJXH GHVFULSWLRQ WKDW LV XQOLNHO\ WR RFFDVLRQ VPRRWK ZHOOH[HFXWHG RSHUDWLRQ RI WKH FOXWFK RQ WKH ILUVW DWWHPSW 2Q WKH RWKHU KDQG WKH OLVWHQHU LV XQOLNHO\ WR UHVSRQG DSSURSULDWHO\ WR HYHQ WKH PRVW DFFXUDWH RI LQVWUXFWLRQV XQGHU VXFK FLUFXPVWDQFHV 7KH LQVWUXFWRU FRXOG WDNH WKH WLPH DQG HIIRUW WR PHDVXUH SUHFLVHO\ WKH UHODWLRQVKLS EHWZHHQ WKH RSHUDWLRQ RI WKH FOXWFK DQG WKH DFFHOHUDWRU DQG WR UHILQH WKH LQVWUXFWLRQ JLYHQ WR WKH SXSLO WR /LIW \RXU OHIW IRRW RII WKH FOXWFK DW D UDWH RI FP SHU VHFRQG ZKLOH GHSUHVVLQJ WKH DFFHOHUDWRU ZLWK \RXU ULJKW IRRW DW D UDWH RI FP SHU VHF :LWKRXW H[WHQVLYH WUDLQLQJ KRZHYHU LW LV XQOLNHO\ WKDW WKH SXSLO ZLOO EH DEOH WR IROORZ WKLV LQVWUXFWLRQ ZLWK DQ\ JUHDWHU GHJUHH RI VXFFHVV WKDQ WKH RULJLQDO LQVWUXFWLRQ ,W LV OLNHO\ WKDW VLPLODU OLPLWV RI GHVFULSWLRQ DQG LQWHUSUHWDWLRQ RFFXU ZKHQ WKH VSHDNHU DQG OLVWHQHU RFFXS\ WKH VDPH VNLQ 7KDW LV WKHUH PD\ EH OLPLWV WR WKH LQIOXHQFH RI VHOI LQVWUXFWLRQ DV ZHOO ([SHUW DWKOHWHV DQG DUWLVWV RIWHQ FDQQRW GHVFULEH WKHLU RZQ
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n VHQVLWLYLW\ WR ORQJWHUP FRQVHTXHQFHV XVLQJ SURFHGXUHV XQOLNHO\ WR JLYH ULVH WR DFFXUDWH GHVFULSWLRQV 7KH SURFHGXUHV ZHUH SDWWHUQHG DIWHU DQ H[SHULPHQW FRQGXFWHG E\ 9DXJKDQ DQG 0LOOHU f GHVLJQHG WR WHVW LQWHUSUHWDWLRQV RI SHUIRUPDQFHV EDVHG XSRQ PD[LPL]DWLRQ RI UHLQIRUFHPHQW GHQVLW\ ,Q WKDW VWXG\ SLJHRQVn NH\SHFNLQJ ZDV PDLQWDLQHG RQ VFKHGXOHV RI UHLQIRUFHPHQW WKDW HVWDEOLVKHG DQ LQYHUVH
UHODWLRQVKLS EHWZHHQ RYHUDOO UHVSRQVH UDWH DQG RYHUDOO UHLQIRUFHPHQW UDWH 2Q WKLV W\SH RI VFKHGXOH UHLQIRUFHPHQW GHQVLW\ LV PD[LPL]HG E\ UHVSRQGLQJ UHODWLYHO\ VORZO\ 8QGHU WKHVH SURFHGXUHV UHLQIRUFHUV DUH VFKHGXOHG DFFRUGLQJ WR D OLQHDU YDULDEOH LQWHUYDO /9,f VFKHGXOH /9, VFKHGXOHV GLIIHU IURP WUDGLWLRQDO YDULDEOH LQWHUYDO 9,f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f DQG IRU D WUDGLWLRQDO 9, V VFKHGXOH GDVKHG OLQHf :KHUHDV WKH 9, IXQFWLRQ DV\PSWRWLFDOO\ DSSURDFKHV WKH SURJUDPPHG UHLQIRUFHPHQW UDWH DFURVV WKH HQWLUH UDQJH RI UHVSRQVH UDWHV WKH /9, IXQFWLRQ LV FRQVWDQW DFURVV DOO EXW WKH ORZHVW UHVSRQVH UDWHV /9, VFKHGXOHV DUH WKXV PRUH IRUJLYLQJ RI GHYLDWLRQV LQ UHVSRQVH UDWH WKDQ WUDGLWLRQDO 9, VFKHGXOHV LQ WKDW UHLQIRUFHPHQW GHQVLW\ LV PD[LPL]HG IRU DOO UHVSRQVH UDWHV JUHDWHU WKDQ RU HTXDO WR WKH LQYHUVH RI WKH VFKHGXOH YDOXH LH UHVSRQVHV SHU PLQ LQ )LJXUH f
5HLQIRUFHUV SHU +RXU 9, V &2167$17 352%$%,/,7< /,1($5 L L L L L L L L L L Ln L f Lnn Â‘nn L Â‘ L Â‘ L Â‘ L f Ln L Â‘ n L rWM 5HVSRQVHV SHU 0LQXWH )LJXUH 0RODU IHHGEDFN IXQFWLRQV IRU OLQHDU DQG FRQVWDQWSUREDELOLW\ 9, V VFKHGXOHV 6HH WH[W IRU GHWDLOVf
$Q LQYHUVH UHODWLRQVKLS EHWZHHQ RYHUDOO UHVSRQVH UDWH DQG RYHUDOO UHLQIRUFHPHQW UDWH FDQ EH HVWDEOLVKHG E\ FRQMRLQWO\ VXEWUDFWLQJ UHLQIRUFHUV IURP WKH VWRUH DFFRUGLQJ WR D IL[HGUDWLR )5f VFKHGXOH 7KH VWRUH FDQ EH GULYHQ LQWR D QHJDWLYH VWDWH IROORZLQJ SHULRGV RI UHODWLYHO\ KLJK UDWH UHVSRQGLQJ WKXV UHTXLULQJ SHULRGV RI UHODWLYHO\ ORZ UDWH UHVSRQGLQJ EHIRUH WKH YDOXH RI WKH VWRUH LV DJDLQ SRVLWLYH DQG UHLQIRUFHPHQW FDQ RFFXU )LJXUH VKRZV WKH UHVXOWLQJ IHHGEDFN IXQFWLRQV ZKHQ UHLQIRUFHUV DUH FRQMRLQWO\ VFKHGXOHG E\ DQ /9, V VFKHGXOH DQG FDQFHOHG HYHU\ RU UHVSRQVHV IURP WRS WR ERWWRP UHVSHFWLYHO\f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rK RU HYHU\ O22A UHVSRQVH 7KH IHHGEDFN IXQFWLRQ SURYLGHV D FRQWLQXXP DORQJ ZKLFK SHUIRUPDQFH FDQ EH HYDOXDWHG ZLWK UHVSHFW WR RYHUDOO UHLQIRUFHPHQW GHQVLW\ 7KH RYHUDOO UHVSRQVH UDWH WKDW PD[LPL]HV RYHUDOO UHLQIRUFHPHQW UDWH ZLOO DOZD\V EH WKH LQYHUVH RI WKH /9, VFKHGXOH YDOXH 6HH )LJXUH f %HFDXVH DQ\ VLJQLILFDQW GHYLDWLRQ IURP WKLV UDWH ZLOO UHVXOW LQ D
5HLQIRUFHUV SHU +RXU )LJXUH 0RODU IHHGEDFN IXQFWLRQV IRU /9, V DQG QHJDWLYH VORSH VFKHGXOHV ZLWK VXEWUDFWLRQ UDWLRV RI DQG
FRUUHVSRQGLQJ GHFUHDVH LQ RYHUDOO UHLQIRUFHPHQW GHQVLW\ GHYLDWLRQV IURP UHLQIRUFHPHQW PD[LPL]DWLRQ DUH UHDGLO\ GHWHFWDEOH 7R WKH H[WHQW WKDW PD[LPL]DWLRQ RI RYHUDOO UHLQIRUFHPHQW GHQVLW\ WDNHV SODFH DW WKH H[SHQVH RI LQFUHDVHV LQ ORFDO GHOD\V WR UHLQIRUFHPHQW WKH QHJDWLYH VORSH SURFHGXUHV VKDUH VRPH FRPPRQ JURXQG ZLWK fVHOIFRQWUROf SURFHGXUHV $V UHVSRQVH UDWH GHFUHDVHV RQ QHJDWLYH VORSH VFKHGXOHV DQG DSSURDFKHV WKH LQYHUVH RI WKH /9, VFKHGXOH YDOXH D JUHDWHU SHUFHQWDJH RI WKH LQWHUYDOV VFKHGXOLQJ UHLQIRUFHPHQW ZLOO EH VKRUWHU WKDQ WKH WLPH EHWZHHQ VXFFHVVLYH UHVSRQVHV RU LQWHUUHVSRQVH WLPHV ,57Vf $V ,57 OHQJWK LQFUHDVHV LH UHVSRQVHV UDWH GHFUHDVHVf WKH SUREDELOLW\ WKDW WKH VWRUH LV SRVLWLYH DOVR LQFUHDVHV $Q\ UHODWLYH DSSUR[LPDWLRQ WR WKH UDWH DW ZKLFK UHLQIRUFHPHQW GHQVLW\ LV PD[LPL]HG ZLOO WKHUHIRUH LQFXU ORQJHU ORFDO GHOD\V WR UHLQIRUFHPHQW 7KXV DV ZLWK fVHOIFRQWUROf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
IURP PD[LPL]DWLRQf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
(;3(5,0(17 0HWKRG 6XEMHFWV 7ZR PDOH DQG WKUHH IHPDOH DGXOW KXPDQV SDUWLFLSDWHG IXOO\ LQ ([SHULPHQW LQ H[FKDQJH IRU PRQH\ 7KH VXEMHFWV ZHUH UHFUXLWHG YLD D FODVVLILHG DGYHUWLVHPHQW LQ D FDPSXV QHZVSDSHU 1RQH RI WKH VXEMHFWV KDG SUHYLRXVO\ RU ZDV FXUUHQWO\ HQUROOHG LQ FRXUVHZRUN LQ EHKDYLRU DQDO\VLV RU OHDUQLQJ WKHRU\ 7KH VXEMHFWV ZHUH LQIRUPHG SULRU WR WKHLU ILUVW VHVVLRQ WKDW WKH\ ZRXOG EH SDLG SHU VHVVLRQ DQG WKDW WKH\ FRXOG UHFHLYH D ERQXV E\ HDUQLQJ SRLQWV ZRUWK HDFK ,Q RUGHU WR HQFRXUDJH IXOO SDUWLFLSDWLRQ FROOHFWLRQ RI ERQXV HDUQLQJV ZDV FRQWLQJHQW XSRQ FRPSOHWLRQ RI WKH VWXG\ 7KH VXEMHFWV ZHUH DOVR LQVWUXFWHG QRW WR EULQJ SHUVRQDO LWHPV HJ IRRG WRROV VPRNLQJ PDWHULDOV SRUWDEOH UDGLRV RU WLPH SLHFHVf LQWR WKH ZRUN VSDFH DQG WKDW YLRODWLRQ RI WKLV UXOH ZRXOG UHVXOW LQ GLVPLVVDO IURP WKH VWXG\ 2YHUDOO HDUQLQJV LQFOXGLQJ ERQXVHVf UDQJHG IURP WR SHU KRXU PHGLDQ SHU KRXUf 6L[ DGGLWLRQDO VXEMHFWV ZHUH UHFUXLWHG IRU WKH VWXG\ EXW GLG QRW SDUWLFLSDWH EH\RQG WKH LQLWLDO FRQGLWLRQ )LYH RI WKHVH VXEMHFWV ZLWKGUHZ IURP WKH VWXG\ DQG DQRWKHU ZDV GLVPLVVHG GXH WR LQVXIILFLHQW EDVHOLQH UHVSRQVH UDWHV 7KH GDWD IURP WKHVH VXEMHFWV ZLOO QRW EH FRQVLGHUHG KHUH
$SSDUDWXV 7KH PDQLSXODQGXP FRQVLVWHG RI WKH ULJKW VLGH RI D FP ; FP GXDO UHVSRQVH SDQHO LOOXPLQDWHG ZLWK UHG OLJKW DQG PRXQWHG FHQWUDOO\ RQ D %56)RULQJHU KXPDQ RSHUDQW SDQHO +7&f $ FRPSXWHU PRQLWRU ZDV VHDWHG RQ WRS RI WKH SDQHO DQG D NH\ERDUG ZDV ORFDWHG LQ IURQW RI WKH SDQHO 'DWD ZHUH UHFRUGHG DQG FRQWLQJHQFLHV FRQWUROOHG E\ D SURJUDP FRPSLOHG LQ 4XLFNEDVLFp UXQQLQJ RQ D &RPSXDGGp PRGHO FRPSXWHU &XPXODWLYH UHVSRQVH UHFRUGV ZHUH FROOHFWHG XVLQJ D *HUEUDQGVp FXPXODWLYH UHFRUGHU 3URFHGXUH 7KH IROORZLQJ LQVWUXFWLRQV ZHUH UHDG WR WKH VXEMHFW SULRU WR WKH ILUVW VHVVLRQ DQG ZHUH GLVSOD\HG RQ WKH FRPSXWHU PRQLWRU SULRU WR WKH VWDUW RI HDFK VHVVLRQ 3OHDVH UHDG FDUHIXOO\
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f V 3DQHO SUHVVLQJ ZDV PDLQWDLQHG RQ D YDULDEOHLQWHUYDO 9,f V VFKHGXOH RI SRLQW GHOLYHU\ 7KH 9, VFKHGXOH FRQVLVWHG RI LQWHUYDOV JHQHUDWHG XVLQJ WKH PHWKRG GHVFULEHG E\ )OHVKLHU DQG +RIIPDQ f 7KH LQWHUYDOV ZHUH UDQGRPL]HG DQG ZHUH WKHQ SUHVHQWHG VHTXHQWLDOO\ ERWK ZLWKLQ DQG DFURVV VHVVLRQV 8SRQ FRPSOHWLRQ RI DQ LQWHUYDO WKH QH[W UHVSRQVH ZRXOG UHVXOW LQ SRLQW GHOLYHU\ 3RLQW GHOLYHU\ ZDV VLJQDOHG E\ D EULHI WRQH DQG E\ LQFUHPHQWLQJ D FRXQWHU WKDW ZDV FRQWLQXRXVO\ GLVSOD\HG RQ WKH FRPSXWHU PRQLWRU WKURXJKRXW WKH EORFN 7KH QH[W LQWHUYDO LQ WKH VHTXHQFH EHJDQ WLPLQJ LPPHGLDWHO\ DIWHU SRLQW GHOLYHU\ /LQHDU YDULDEOH LQWHUYDO /9,f V &RQGLWLRQV ZHUH WKH VDPH DV LQ WKH 9, V FRQGLWLRQ H[FHSW WKDW XSRQ FRPSOHWLRQ RI DQ LQWHUYDO D SRLQW ZDV VWRUHG LQ D EDQN RI DYDLODEOH UHLQIRUFHUV DQG WKH QH[W LQWHUYDO EHJDQ WLPLQJ LPPHGLDWHO\ ,I D SUHVV RFFXUUHG DQG WKH YDOXH RI WKH VWRUH ZDV SRVLWLYH RQH SRLQW ZDV GHOLYHUHG DQG RQH ZDV VXEWUDFWHG IURP WKH VWRUH ,Q WKH DEVHQFH RI UHVSRQGLQJ WKH QXPEHU RI DYDLODEOH UHLQIRUFHUV LQ WKH
7DEOH 7KH VHTXHQFH RI FRQGLWLRQV DQG WKH QXPEHU RI VHVVLRQV FRQGXFWHG XQGHU HDFKf IRU DOO VXEMHFWV XQGHU YDULDEOH LQWHUYDO 9,f OLQHDU YDULDEOH LQWHUYDO /9,f DQG QHJDWLYH VORSH 16f VFKHGXOHV RI SRLQW GHOLYHU\ 6LOO 6 6XEMHFW 6 6 6 9, f 9, f 9,f 9, f 9, f /9, f /9, f /9, f &21-7 f /9, f 16 f 16 f 16 f 9, f 16 f 16 f 16 f 16 f /9, f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f 16 f
VWRUH ZRXOG FRQWLQXH WR HVFDODWH ZLWK WKH FRPSOHWLRQ RI HDFK VXFFHVVLYH LQWHUYDO 6XEMHFWV ZHUH QRW LQIRUPHG RI WKH YDOXH RI WKH VWRUH ZKLFK FDUULHG DFURVV EORFNV DQG VHVVLRQV QRU RI WKH FKDQJH LQ FRQGLWLRQV 1HJDWLYH VORSH 16f )5 Q ,Q WKLV FRQGLWLRQ SRLQWV ZHUH DGGHG WR WKH VWRUH DFFRUGLQJ WR DQ /9, V VFKHGXOH DV GHVFULEHG DERYH EXW HYHU\ mWK UHVSRQVH VXEWUDFWHG RQH SRLQW IURP WKH VWRUH ,I SRLQWV ZHUH VXEWUDFWHG PRUH IUHTXHQWO\ WKDQ WKH\ ZHUH VHW XS E\ WKH /9, VFKHGXOH WKH VWRUH FRXOG EH GULYHQ LQWR D QHJDWLYH VWDWH %RWK WKH YDOXH RI WKH VWRUH DQG WKH SRVLWLRQ ZLWKLQ WKH )5 ZHUH FDUULHG DFURVV VHVVLRQV 7KH VXEWUDFWLRQ UDWLRV IRU LQGLYLGXDO VXEMHFWV DUH OLVWHG LQ 7DEOH 7KHVH YDOXHV ZHUH EDVHG XSRQ WKH PHDQ UHVSRQVH UDWHV IURP WKH ODVW ILYH VHVVLRQV LQ WKH SUHFHGLQJ FRQGLWLRQ DQG ZHUH VHOHFWHG VXFK WKDW WKH RYHUDOO UDWH RI SRLQW GHOLYHU\ ZRXOG EH UHGXFHG E\ RQH KDOI LI WKH VWHDG\ VWDWH UHVSRQVH UDWHV RI WKH SUHFHGLQJ FRQGLWLRQ SUHYDLOHG ,Q WKH ILUVW QHJDWLYHVORSH FRQGLWLRQ IRU 6XEMHFW IRU H[DPSOH WKH VXEWUDFWLRQ UDWLR YDOXH ZDV HTXDO WR WKH PHDQ UHVSRQVHV SHU PLQ IURP WKH ODVW VHVVLRQV RI WKH /9, V FRQGLWLRQ ,I WKLV VXEMHFW FRQWLQXHG WR UHVSRQG DW UHVSRQVHV SHU PLQ LQ WKH 16 FRQGLWLRQ WKHQ RQH SRLQW SHU PLQXWH ZRXOG EH VXEWUDFWHG IURP WKH VWRUH DFFRUGLQJ WR WKH )5 ZKLOH WKH /9, V VFKHGXOH ZDV VLPXOWDQHRXVO\ DGGLQJ SRLQWV SHU PLQ WR WKH VWRUH 7KH UHVXOWLQJ QHW UHLQIRUFHPHQW UDWH ZRXOG EH DSSUR[LPDWHO\ RQH SRLQW SHU PLQXWH RU DSSUR[LPDWHO\ KDOI WKH VWHDG\VWDWH UDWH RI SRLQW GHOLYHU\ IURP WKH /9, FRQGLWLRQ 7KH VXEWUDFWLRQ UDWLRV RI 6XEMHFWV DQG ZHUH GHWHUPLQHG VOLJKWO\ GLIIHUHQWO\ 7KH LQLWLDO VXEWUDFWLRQ UDWLR ZDV VHOHFWHG VXFK WKDW WKH RYHUDOO UDWH RI SRLQW GHOLYHU\ ZRXOG EH SHU PLQXWH LQVWHDG RI SHU PLQXWH %RWK RI WKHVH VXEMHFWV VKRZHG OLWWOH
VHQVLWLYLW\ WR WKH QHJDWLYH VORSH VFKHGXOHV DIWHU VHYHUDO FRQGLWLRQV 7R SURGXFH GHFUHDVHV LQ UHVSRQVH UDWHV LQ WKH 16 DQG 16 FRQGLWLRQV IRU 6XEMHFW DQG LQ WKH 16 DQG 16 FRQGLWLRQV IRU 6XEMHFW VXEWUDFWLRQ UDWLRV WKDW UHGXFHG WKH UDWH RI SRLQW GHOLYHU\ WR DSSUR[LPDWHO\ SHU PLQXWH ZHUH XVHG ,Q WKH ILQDO QHJDWLYH VORSH 16 f FRQGLWLRQ IRU 6XEMHFW WKH WDUJHWHG UDWH RI SRLQW GHOLYHU\ ZDV SHU PLQXWH &RQMRLQW 9, V ,57 Vf5DQGRP UDWLR ,57 Vf &21-79755f )RU 6XEMHFW WKH UDWH RI UHVSRQGLQJ IROORZLQJ WKH LQLWLDO VHVVLRQ H[SRVXUH WR WKH 9, FRQGLWLRQ ZDV LQVXIILFLHQW WR SURJUHVV WR WKH UHPDLQLQJ FRQGLWLRQV 7R UHFWLI\ WKLV WKH VXEMHFW ZDV H[SRVHG WR D FRPSOH[ VFKHGXOH RI SRLQW GHOLYHU\ LQ ZKLFK LQWHUUHVSRQVH WLPHV ,57Vf OHVV WKDQ V ZHUH UHLQIRUFHG DFFRUGLQJ WR UDQGRP UDWLR 55f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
ZDV UHLPSOHPHQWHG 7R PDLQWDLQ RYHUDOO VHVVLRQ HDUQLQJV FRPSDUDEOH WR WKH RWKHU FRQGLWLRQV VHVVLRQV ZHUH FRPSULVHG RI EORFNV WKDW WHUPLQDWHG DIWHU PLQ RU SRLQW GHOLYHULHV ZKLFKHYHU FDPH ILUVW 3URFHGXUDO LUUHJXODULWLHV 7KHUH ZHUH D IHZ GHYLDWLRQV IURP WKH SURFHGXUH GXULQJ WKH FRXUVH RI WKH H[SHULPHQW 6XEMHFW GLG QRW UHVSRQG GXULQJ WKH ILUVW VHVVLRQ 3ULRU WR WKH VHFRQG VHVVLRQ WKH H[SHULPHQWHU UHUHDG WKH LQVWUXFWLRQV WR WKH VXEMHFW DQG JHVWXUHG WRZDUG WKH DSSURSULDWH UHVSRQVH SDQHO )RU 6XEMHFWV DQG WKH ILQDO FRQGLWLRQ ZDV FRQFOXGHG DUELWUDULO\ )RU 6XEMHFWV DQG WKH ILQDO FRQGLWLRQ HQGHG GXH WR VFKHGXOH FRQIOLFWV 6XEMHFW OHIW WRZQ GXULQJ D VHPHVWHU EUHDN DQG GLG QRW UHWXUQ WR WKH H[SHULPHQW 5HVXOWV 0RODU )HHGEDFN )XQFWLRQV )LJXUH VKRZV WKH PHDQ UDWH RI SRLQW GHOLYHU\ YHUVXV WKH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV XQGHU DOO FRQGLWLRQV IRU HDFK VXEMHFW 7KHVH SRLQWV KDYH EHHQ VXSHULPSRVHG XSRQ WKH IHHGEDFN IXQFWLRQV IRU WKH UHVSHFWLYH VFKHGXOHV 7KH XSSHUPRVW FXUYH LQ HDFK JUDSK LV WKH PRODU IHHGEDFN IXQFWLRQ IRU WKH /9, FRQGLWLRQ DQG WKH GDVKHG FXUYH EHORZ LW LV WKH PRODU IHHGEDFN IXQFWLRQ IRU WKH 9, FRQGLWLRQ 7KH GLDPRQGV DQG WKH WULDQJOHV LQ )LJXUH VKRZ WKH GDWD REWDLQHG XQGHU WKH /9, DQG WKH 9, FRQGLWLRQV UHVSHFWLYHO\ 5DWH RI SRLQW GHOLYHU\ LQFUHDVHG IROORZLQJ WKH WUDQVLWLRQ WR WKH /9, IURP WKH 9, FRQGLWLRQV IRU DOO VXEMHFWV DV LQGLFDWHG LQ WKH ILJXUH E\ D VOLJKW YHUWLFDO GLVSODFHPHQW RI WKH GLDPRQGV UHODWLYH WR WKH SRVLWLRQV RI WKH WULDQJOHV WKLV UHVXOW
)LJXUH 0HDQ UDWH RI SRLQW GHOLYHU\ YHUVXV WKH PHDQ UHVSRQVH UDWHV IURP WKH ODVW VHVVLRQV RI WKH 9, V /9, V DQG 16 FRQGLWLRQV IRU HDFK VXEMHFW 7KH GDWD DUH VXSHULPSRVHG RQ WKH PRODU IHHGEDFN IXQFWLRQV IRU HDFK FRQGLWLRQ 1XPEHUV QHDU WKH DEVFLVVDH LQGLFDWH WKH VXEWUDFWLRQ UDWLRV RI WKH 16 FRQGLWLRQV 1RWH LQGLYLGXDOO\ VFDOHG D[HV
RLQWV SHU +RXU 6 6 6 6 fWRR 5HVSRQVHV SHU
LV LQ DFFRUG ZLWK WKH SURJUDPPHG FKDQJHV LQ WKH FRQWLQJHQFLHV DFURVV WKH WZR FRQGLWLRQV DQG RFFXUUHG ZKHWKHU UHVSRQVH UDWHV LQFUHDVHG 6XEMHFWV t f GHFUHDVHG 6XEMHFW f RU UHPDLQHG WKH VDPH 6XEMHFW f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
LQGLFDWH D UHGXFWLRQ LQ UHVSRQVH UDWH DQG D FRUUHVSRQGLQJ LQFUHDVH LQ UHLQIRUFHPHQW UDWH )RU H[DPSOH WKH ODUJHVW DEVROXWH FKDQJH LQ UHVSRQVH UDWH RFFXUUHG ZLWK 6XEMHFW GXULQJ H[SRVXUH WR WKH ILUVW QHJDWLYHVORSH FRQGLWLRQ 16 f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f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t 16 IRU 6XEMHFW f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t VKRZ WKH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV :KHUHDV RQO\ VWHDG\VWDWH GDWD ZHUH SRUWUD\HG LQ )LJXUH WKHVH ILJXUHV VKRZ WUDQVLWLRQ GDWD DQG WUHQGV LQ UHVSRQVH DQG UHLQIRUFHPHQW UDWHV
)LJXUH 7KH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 1RWH LQGLYLGXDOO\ VFDOHG D[HV
5HVSRQVHV SHU 0LQXWH fÂ§N fÂ§ fÂ§r WR .f WR UR .! 8O R R .f FQ R UR 2 8L R FQ R 8L R R &R FQ 3RLQWV SHU 0LQXWH 2 f 5HVSRQVHV 3RLQWV 6HVVLRQV 5HVSRQVHV SHU 0LQXWH .f&QA-2I1RFU!nA-2.fMLnA-2.f&QnAL RFQRFQRXLRFMLRDLRXLRFQRXL 3RLQWV SHU 0LQXWH WR 02
)RU 6XEMHFW WRS SDQHO )LJ f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f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f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
)LJXUH 7KH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 1RWH LQGLYLGXDOO\ VFDOHG D[HV
6HVVLRQV 5HVSRQVHV SHU 0LQXWH fIRUR0IRXL 1-8LnYMRURFQnJRURP A R RFQRFQRXLRFQRFQRXLR Â•8Â•/QFQ/Q 3RLQWV SHU 0LQXWH XR KR 2 f 5HVSRQVHV 3RLQWV 6HVVLRQV 5HVSRQVHV SHU 0LQXWH 1! 2, n8 2 1! FQ R RPRXLRFQRFQ 3RLQWV SHU 0LQXWH XR UR WR WR 8-
8QGHU WKH QHJDWLYHVORSH FRQGLWLRQV UHVSRQVH UDWHV DQG UDWHV RI SRLQW GHOLYHU\ GHFUHDVHG ZLWK HDFK VXFFHVVLYH GHFUHDVH LQ WKH VXEWUDFWLRQ UDWLR 8QGHU WKH 16 FRQGLWLRQ WKH UHVSRQVH UDWHV GHFUHDVHG VOLJKWO\ UHODWLYH WR WKH UDWHV RI WKH SUHFHGLQJ /9, FRQGLWLRQ DQG DV D UHVXOW WKHUH ZDV D ODUJH GHFUHDVH LQ UDWH RI SRLQW GHOLYHU\ ,Q DGGLWLRQ EHWZHHQVHVVLRQ YDULDELOLW\ LQ UDWHV RI SRLQW GHOLYHU\ LQFUHDVHG XQGHU WKH 16 FRQGLWLRQ ,Q WKH VXEVHTXHQW FRQGLWLRQ 16 f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
7KLV VXEMHFW ZDV H[SRVHG WR IRXU QHJDWLYH VORSH FRQGLWLRQV GXULQJ ZKLFK UHVSRQVH UDWHV GHFUHDVHG ZLWKLQ DQG DFURVV FRQGLWLRQV 5DWHV RI SRLQW GHOLYHU\ GHFUHDVHG IROORZLQJ WKH WUDQVLWLRQ IURP RQH QHJDWLYH VORSH FRQGLWLRQ WR WKH QH[W EXW LQFUHDVHG ZLWKLQ HDFK FRQGLWLRQ 7KH ODUJHVW DEVROXWH FKDQJH LQ UHVSRQVH UDWHV RFFXUUHG GXULQJ WKH WKLUG QHJDWLYH VORSH FRQGLWLRQ 16 f DV D UHVXOW WKH VWHDG\ VWDWH UDWH RI SRLQW GHOLYHU\ H[FHHGHG WKRVH RI WKH SUHYLRXV WZR FRQGLWLRQV 16 t 16 f 5HVSRQVH UDWHV GHFUHDVHG IXUWKHU LQ WKH ILQDO QHJDWLYH VORSH FRQGLWLRQ 16 f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
5HVSRQVHV SHU 0LQXWH 6 )LJXUH 7KH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV IRU 6XEMHFW 3RLQWV SHU 0LQXWH
7KH ODUJHVW DEVROXWH GHFUHDVH LQ UHVSRQVH UDWH RFFXUUHG ZLWKLQ WKH ILQDO FRQGLWLRQ 16 f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f LV IURP 6HVVLRQ GXULQJ WKH ILUVW H[SRVXUH WR WKH 9, FRQGLWLRQ 7KH VORSH RI WKH OLQH LV VKDOORZ DQG URXJKO\ FRQVWDQW DFURVV WKH PLQ EORFN LQGLFDWLQJ D ORZ EXW VWHDG\ UDWH RI UHVSRQGLQJ 7KH VHJPHQW LQ 3DQHO LV IURP 6HVVLRQ GXULQJ WKH VHFRQG H[SRVXUH WR WKH 9, FRQGLWLRQ IROORZLQJ WKH &RQMRLQW FRQGLWLRQf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f RFFDVLRQDO SDXVHV
&XPXORWLYH 5HVSRQVHV 6 9, V 9, V 6HVVLRQ 6HVVLRQ /9, V 16 6HVVLRQ 6HVVLRQ 16 6HVVLRQ 0LQXWHV 16 6HVVLRQ 16 6HVVLRQ )LJXUH 3RUWLRQV RI UHSUHVHQWDWLYH FXPXODWLYH UHVSRQVH UHFRUGV IURP VHVVLRQV FRQGXFWHG XQGHU 9, /9, DQG 16 FRQGLWLRQV IRU 6XEMHFW (DFK SDQHO LV WKH VHJPHQW RI WKH FXPXODWLYH UHFRUG IURP WKH VHFRQG PLQ EORFN RI WKH VHVVLRQ
LQWHUUXSW SHULRGV RI KLJKUDWH UHVSRQGLQJ ,Q WKH VXEVHTXHQW FRQGLWLRQ 16 6HVVLRQ f WKH KLJKUDWH SDWWHUQV ZHUH VLPLODU EXW SDXVLQJ ZDV PRUH IUHTXHQW 3DQHO 16 6HVVLRQ f VKRZV UHVSRQVH SDWWHUQLQJ WKDW LV FKDUDFWHULVWLF RI WHUPLQDO QHJDWLYH VORSH VFKHGXOH SHUIRUPDQFH SDXVHV DOWHUQDWH ZLWK SHULRGV RI KLJKUDWH UHVSRQGLQJ 7KH KLJK UDWH UHVSRQVH UXQV IUHTXHQWO\ EHJDQ ZLWK D VHULHV RI SRLQW GHOLYHULHV DV WKH VWRUH ZDV GHSOHWHG DQG W\SLFDOO\ FRQWLQXHG IRU PDQ\ UHVSRQVHV WKHUHDIWHU 2FFDVLRQDOO\ SRLQW GHOLYHU\ RFFXUUHG LQ WKH ODWWHU SRUWLRQ RI WKHVH UHVSRQVH UXQV 7KLV SDXVHUXQ SDWWHUQ LV HYHQ PRUH SURQRXQFHG DQG FRQVLVWHQW LQ 3DQHO 16 6HVVLRQ f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
)LJXUH 6FDWWHU SORWV RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ DFURVV VHVVLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 1RWH LQGLYLGXDOO\ VFDOHG DQG ORJDULWKPLF D[HV
6HVVLRQV ,57 3UHFHGLQJ 3RLQW 'HOLYHU\ 6HFf R L R 2 1f R R 0 .f 6 RR ,57 %HIRUH 3RLQW 'HOLYHU\ 6HFf R fÂ§n } L R R R 1! 1f 21f fÂ§r1f 1f1f 6
)LJXUH 6FDWWHU SORWV RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ DFURVV VHVVLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 1RWH LQGLYLGXDOO\ VFDOHG DQG ORJDULWKPLF D[HV
6HVVLRQV ,57 %HIRUH 3RLQW 'HOLYHU\ 6HFf 6 6HVVLRQV ,57 3UHFHGLQJ 3RLQW 'HOLYHU\ 6HFf 86
,57V 3UHFHGLQJ 3RLQW 'HOLYHU\ 7KH SDXVHUXQ SDWWHUQ RI UHVSRQGLQJ XQGHU QHJDWLYH VORSH FRQGLWLRQV DSSDUHQW LQ WKH FXPXODWLYH UHFRUGV VXJJHVWHG WKDW WKH ,57 SUHFHGLQJ UHLQIRUFHPHQW PD\ EH D XVHIXO PHDVXUH WR WUDFN DFURVV FRQGLWLRQV )LJXUHV DQG DUH VFDWWHU SORWV RI WKH LQWHUUHVSRQVH WLPHV SUHFHGLQJ SRLQW GHOLYHU\ DFURVV VHVVLRQV IRU DOO VXEMHFWV H[FHSW 6XEMHFW WKH ILUVW VXEMHFW UXQ LQ WKH H[SHULPHQW ZDV WKH VXEMHFW ZKRVH GDWD VXJJHVWHG WKH SUHVHQW PHDVXUHf &RQVLVWHQW ZLWK WKH SDXVHUXQ SDWWHUQV RI WKH FXPXODWLYH UHFRUGV E\ WKH FRQFOXVLRQ RI WKH QHJDWLYHVORSH FRQGLWLRQV WKH GLVWULEXWLRQV RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ ZHUH ELPRGDO IRU WKUHH RI WKHVH IRXU VXEMHFWV t f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f 5DWHV RI SRLQW GHOLYHU\ KRZHYHU ZHUH QRW PD[LPL]HG LQ DQ\ RI WKH QHJDWLYH VORSH FRQGLWLRQV VXJJHVWLQJ WKDW VHQVLWLYLW\ WR WKH LQYHUVH UHODWLRQVKLS EHWZHHQ UHVSRQVH UDWH DQG UHLQIRUFHPHQW UDWH ZDV OLPLWHG 7KH UDWHV
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t f IRU ZKLFK WKLV PHDVXUH ZDV UHFRUGHG ZHUH ELPRGDO ZLWK RQH FOXVWHU RI VKRUW ,57V HJ OHVV WKDQ Vf DQG DQRWKHU FOXVWHU RI ORQJHU ,57V HJ JUHDWHU WKDQ Vf )RU 6XEMHFW WKH GLVWULEXWLRQV RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ ZHUH QRW ELPRGDO EXW WKH IUHTXHQF\ RI UHODWLYHO\ ORQJ ,57V LQFUHDVHG XQGHU WKH QHJDWLYH VORSH FRQGLWLRQV 6RPH FRPSHOOLQJ HYLGHQFH RI D OLPLWDWLRQ LQ VHQVLWLYLW\ WR WKH LQYHUVH UHODWLRQVKLS EHWZHHQ RYHUDOO UDWH RI SRLQW GHOLYHU\ DQG RYHUDOO UHVSRQVH UDWH FRPHV IURP VHVVLRQV LQ
ZKLFK XQXVXDOO\ ORQJ SDXVHV ZHUH IROORZHG E\ VXEVWDQWLDO LQFUHDVHV LQ UHLQIRUFHPHQW UDWH 2Q IRXU RFFDVLRQV DFURVV VXEMHFWV WKH HQG RI D PLQ EORFN ZDV SUHFHGHG E\ D SDUWLFXODUO\ OHQJWK\ SDXVH HJ LQ H[FHVV RI PLQf 7KLV RFFXUUHG WZLFH IRU 6XEMHFW 6HVVLRQV t f DQG RQFH HDFK IRU 6XEMHFWV DQG 6HVVLRQV DQG UHVSHFWLYHO\f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t f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f ,QWHUHVWLQJO\ WKHVH H[SHULHQFHV GLG QRW DSSHDU WR LQIOXHQFH LPPHGLDWHO\ WKH RYHUDOO UDWH RU SDWWHUQ RI UHVSRQGLQJ LQ WKH VXEVHTXHQW VHVVLRQ RU EORFN 6HQVLWLYLW\ WR WKH FRQWLQJHQFLHV RI WKH QHJDWLYH VORSH VFKHGXOH ZDV DSSDUHQW HDUOLHU LQ WKH VHTXHQFH RI FRQGLWLRQV IRU 6XEMHFW WKDQ IRU WKH RWKHU VXEMHFWV 7KH UHVSRQVH UDWHV
RI 6XEMHFW GHFUHDVHG VXEVWDQWLDOO\ GXULQJ 6HVVLRQ WKH WKLUG VHVVLRQ RI WKH ILUVW QHJDWLYH VORSH FRQGLWLRQ 6HH )LJXUH f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
KRZHYHU VKRZ WKLV SDWWHUQ UREXVWO\ 7KH UHODWLRQVKLS LV QRW DV SURQRXQFHG LQ WKH GDWD IURP WKH QHJDWLYHVORSH FRQGLWLRQV IRU WKH UHPDLQLQJ VXEMHFWV /LNH WKH KXPDQV LQ WKH SUHVHQW VWXG\ WKH SHUIRUPDQFH RI QRQKXPDQV RQ QHJDWLYH VORSH VFKHGXOHV LV QRW ZHOO FKDUDFWHUL]HG E\ UHLQIRUFHPHQW GHQVLW\ PD[LPL]DWLRQ )LJXUH UHSULQWHG IURP 9DXJKDQ f VKRZV WKH VWHDG\VWDWH UHVSRQVH UDWHV IURP WKH QHJDWLYH VORSH FRQGLWLRQV LQ 9DXJKDQ DQG 0LOOHUnV f H[SHULPHQW VXSHULPSRVHG XSRQ WKH PRODU IHHGEDFN IXQFWLRQV IRU WKH UHVSHFWLYH VFKHGXOHV 7KH UHVSRQVH UDWHV GLIIHUHG VXEVWDQWLDOO\ IURP WKH LQYHUVH RI WKH /9, YDOXHV DQG WKH UHLQIRUFHPHQW UDWHV IUHTXHQWO\ DSSURDFKHG SHU PLQXWH 'HVSLWH VXEVWDQWLDO H[SHULHQFH ZLWK WKHVH FRQWLQJHQFLHV ORZ UHVSRQVH UDWHV ZHUH QRW REWDLQHG DQG RYHUDOO UHLQIRUFHPHQW GHQVLW\ ZDV QRW PD[LPL]HG 7KHVH UHVXOW KDYH EHHQ VXEVHTXHQWO\ UHSOLFDWHG ZLWK UDWV (WWLQJHU 5HLG t 6WDGGRQ 5HHG t 6FKDFKWPDQ f &RPSDULQJ WKH VWHDG\VWDWH GDWD SUHVHQWHG LQ )LJXUH WR WKRVH SUHVHQWHG LQ )LJXUH IURP WKH 9DXJKDQ DQG 0LOOHU f VWXG\ VXJJHVWV WKDW WKH SHUIRUPDQFHV RI KXPDQV DUH VRPHZKDW VLPLODU WR WKRVH RI SLJHRQV ,Q WKH 9DXJKDQ DQG 0LOOHU f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
5(,1)25&(0(176 3(5 +285 )LJXUH 5HLQIRUFHPHQW UDWHV YHUVXV UHVSRQVH UDWHV RI SLJHRQV XQGHU 16 ILOOHG FLUFOHV DQG VROLG OLQHVf DQG \RNHG /9, RSHQ FLUFOHV DQG GDVKHG OLQHVf FRQGLWLRQV IURP 9DXJKDQ DQG 0LOOHU f 7KH GDWD KDYH EHHQ SORWWHG RQ WKH PRODU IHHGEDFN IXQFWLRQV IRU WKH UHVSHFWLYH FRQGLWLRQV )URP f&KRLFH DQG WKH 5HVFRUOD:DJQHU PRGHOff E\ : 9DXJKDQ -U LQ 0 / &RPPRQV 5 +HUUQVWHLQ t + 5DFKOLQ (GVf 4XDQWLWDWLYH DQDO\VHV RI EHKDYLRU 9RO 0DWFKLQJ DQG PD[LPL]LQJ DFFRXQWV SS f &DPEULGJH 0$ %DOOLQJHU &RS\ULJKW E\ /DZUHQFH (UOEDXP $VVRFLDWHV ,QF 5HSULQWHG E\ SHUPLVVLRQ
WKH SUHVHQW VWXG\ D ZLWKLQVXEMHFW GHVLJQf WKH GDWD LQ )LJXUH VXJJHVW WKDW UDWHV RI UHVSRQGLQJ DQG UHLQIRUFHPHQW ZHUH LQYHUVHO\ UHODWHG WR WKH VXEWUDFWLRQ UDWLR YDOXH 7KLV ZDV IUHTXHQWO\ REVHUYHG ZLWK WKH KXPDQ VXEMHFWV DV ZHOO 6HH )LJXUH f 7KHUH ZHUH WZR LQVWDQFHV LQ WKH SUHVHQW VWXG\ KRZHYHU WKDW ZHUH QRW FRQVLVWHQW ZLWK WKLV SDWWHUQ ,Q WKH 16 FRQGLWLRQ IRU 6 DQG WKH 16 FRQGLWLRQ IRU 6 UHVSRQVH UDWHV GHFUHDVHG VXIILFLHQWO\ WR SURGXFH VXEVWDQWLDO LQFUHDVHV LQ WKH UDWHV RI SRLQW GHOLYHU\ UHODWLYH WR WKRVH RI FRQGLWLRQV ZLWK ODUJHU VXEWUDFWLRQ UDWLRV 6HH )LJXUH f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f VWXG\ LW LV OHVV OLNHO\ WKDW VXFK SDWWHUQV ZRXOG KDYH GHYHORS EHFDXVH D PLQLPXP V LQWHUUHLQIRUFHPHQW LQWHUYDO ZDV LPSRVHG 7KDW LV LI WKHUH ZDV PRUH WKDQ RQH UHLQIRUFHU VWRUHG SHFNV ZLWKLQ V DIWHU UHLQIRUFHU SUHVHQWDWLRQ ZRXOG QRW SURGXFH DQRWKHU UHLQIRUFHU 7KLV FRQWLQJHQF\ ZDV RPLWWHG IURP WKH SUHVHQW VWXG\ WR HQKDQFH WKH LQYHUVH UHODWLRQVKLS EHWZHHQ UHVSRQVH UDWHV DQG UHLQIRUFHPHQW UDWHV WKHUHE\ SURYLGLQJ D PRUH VWULQJHQW WHVW RI UHLQIRUFHPHQW GHQVLW\ PD[LPL]DWLRQ
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f LQWURGXFHG D K\SHUEROLF IXQFWLRQDO UHODWLRQ WR DFFRXQW IRU VWHDG\ VWDWH UHVSRQVH UDWHV RQ 9, VFKHGXOHV ,W LV EDVHG XSRQ WKH SUHPLVH RI WKH PDWFKLQJ ODZ ZKLFK VWDWHV WKDW WKH UHODWLYH UDWH RI UHVSRQGLQJ HTXDOV WKH UHODWLYH UDWH RI UHLQIRUFHPHQW SURGXFHG E\ WKDW EHKDYLRU 7KH H[SUHVVLRQ WDNHV WKH JHQHUDO IRUP RI % NUU Uff f ZKHUH % HTXDOV UHVSRQVH UDWH U HTXDOV WKH REWDLQHG UDWH RI UHLQIRUFHPHQW SURGXFHG E\ % N HTXDOV WKH PD[LPXP YDOXH RI % DQG U HTXDOV WKH UHLQIRUFHPHQW UDWH FRUUHVSRQGLQJ WR % %UDGVKDZ HW DO f U LV WKHRUHWLFDOO\ HTXDO WR WKH UDWH RI XQVFKHGXOHG UHLQIRUFHPHQW SURGXFHG E\ EHKDYLRU RWKHU WKDQ % :LWK WKH DLG RI WKH WZR IUHH
SDUDPHWHUV +HUPVWHLQ f ZDV DEOH WR ILW WKLV FXUYH WR GDWD REWDLQHG E\ &DWDQLD DQG 5H\QROGV f ZLWK SLJHRQV UHVSRQGLQJ RQ D YDULHW\ RI 9, VFKHGXOHV 7KH GDWD IURP WKH QHJDWLYH VORSH FRQGLWLRQV LQ WKH 9DXJKDQ DQG 0LOOHU f H[SHULPHQW VXJJHVW WKDW RYHUDOO UHVSRQVH UDWH PLJKW PDWFK RYHUDOO UHLQIRUFHPHQW UDWH 7KH ILOOHG FLUFOHV LQ )LJXUH VKRZ WKH REWDLQHG UHVSRQVH UDWHV YHUVXV WKH REWDLQHG UHLQIRUFHPHQW UDWHV IURP QHJDWLYH VORSH FRQGLWLRQV $OWKRXJK QRQH RI WKH SLJHRQV ZDV H[SRVHG WR HQRXJK FRQGLWLRQV WR GLVFHUQ D K\SHUEROLF UHODWLRQVKLS UHVSRQVH UDWHV YDULHG GLUHFWO\ ZLWK UHLQIRUFHPHQW UDWHV IRU RI WKH ELUGV DQ RXWFRPH WKDW LV FRQVLVWHQW ZLWK DFFRXQWV EDVHG XSRQ UHODWLYH UHLQIRUFHPHQW UDWH +HUPVWHLQ 3UHOHF f 7KLV RXWFRPH VXJJHVWV WKDW WKH SLJHRQVn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n EHKDYLRU LV LQGHHG VHQVLWLYH WR WKH LQYHUVH UHODWLRQVKLS EHWZHHQ UHVSRQGLQJ DQG UHLQIRUFHPHQW 7KH UHVXOWV RI VXFK D PDQLSXODWLRQ FDQ EH VHHQ LQ )LJXUH ,Q WKH 9DXJKDQ DQG 0LOOHU f VWXG\ WKH QHJDWLYH VORSH FRQGLWLRQV DOWHUQDWHG ZLWK /9, FRQGLWLRQV LQ ZKLFK WKH RYHUDOO UHLQIRUFHPHQW UDWHV HTXDOHG WKRVH RI WKH SUHFHGLQJ QHJDWLYHVORSH FRQGLWLRQV 7KH GDWD IURP WKH /9, FRQGLWLRQV RSHQ FLUFOHVf KDYH EHHQ VXSHULPSRVHG XSRQ WKH PRODU IHHGEDFN IXQFWLRQV IRU WKRVH VFKHGXOHV GDVKHG OLQHVf LQ WKH ILJXUH $OWKRXJK QRW D SDUWLFXODUO\ UREXVW HIIHFW WKH UHVSRQVH UDWHV WHQGHG WR EH KLJKHU LQ WKH
/9, FRQGLWLRQV WKDQ LQ WKH DVVRFLDWHG QHJDWLYH VORSH FRQGLWLRQV LQGLFDWLQJ VRPH GHJUHH RI VHQVLWLYLW\ WR WKH QHJDWLYH VORSH FRQWLQJHQFLHV 7KH WHQGHQF\ IRU KLJKHU UHVSRQVH UDWHV XQGHU WKH \RNHG /9, VFKHGXOHV WKDQ XQGHU QHJDWLYH VORSH VFKHGXOHV ZLWK FRPSDUDEOH UHLQIRUFHPHQW UDWHV KDV EHHQ UHSOLFDWHG LQ D VHULHV RI H[SHULPHQWV E\ 5HHG DQG 6FKDFKWPDQ f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t 6FKDFKWPDQ t UHVSHFWLYHO\f ,Q WKH ODWWHU VWXG\ WKH UROHV RI WKH VXEMHFWV ZHUH ODWHU UHYHUVHG VXFK WKDW HDFK VXEMHFW KDG H[SHULHQFH DV ERWK D PDVWHU VXEMHFW DQG D \RNHG VXEMHFW ,Q ERWK H[SHULPHQWV UDWV LQ WKH \RNHG JURXSV WHQGHG WR UHVSRQG DW VOLJKWO\ KLJKHU UDWHV WKDQ WKH PDVWHU UDWV WKDW H[SHULHQFHG WKH QHJDWLYH VORSH VFKHGXOHV KRZHYHU EHWZHHQ VXEMHFW GLIIHUHQFHV LQ SHUIRUPDQFH PDGH LQWHUSUHWDWLRQ RI WKH UHVXOWV WHQXRXV DW EHVW ,Q DQRWKHU H[SHULPHQW E\ 5HHG DQG 6FKDFKWPDQ f WKH FRPSDULVRQ EHWZHHQ SHUIRUPDQFHV XQGHU QHJDWLYHVORSH DQG \RNHG /9, VFKHGXOHV ZDV DFFRPSOLVKHG LQ WKH FRQWH[W RI D ZLWKLQVXEMHFW GHVLJQ )RRG ZDV GHOLYHUHG DFFRUGLQJ WR D QHJDWLYH VORSH VFKHGXOH LQ RQH FRPSRQHQW RI D PXOWLSOH VFKHGXOH RI UHLQIRUFHPHQW DQG DFFRUGLQJ WR D
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f V DQG OLQHDU YDULDEOH LQWHUYDO /9,f V 7KH FRQGLWLRQV ZHUH WKH VDPH DV WKRVH GHVFULEHG DERYH IRU ([SHULPHQW 1HJDWLYH VORSH 16f )5 Q 7KH FRQWLQJHQFLHV RI WKH QHJDWLYH VORSH FRQGLWLRQV ZHUH WKH VDPH DV WKRVH GHVFULEHG LQ ([SHULPHQW $V EHIRUH WKH GHWHUPLQDWLRQ ZDV EDVHG XSRQ WKH PHDQ UHVSRQVH UDWHV IURP WKH ILQDO VHVVLRQV RI WKH /9, FRQGLWLRQ WKH YDOXH ZDV VHOHFWHG VXFK WKDW WKH RYHUDOO UDWH RI SRLQW GHOLYHU\ ZRXOG HTXDO SRLQWV SHU PLQ LI WKH UHVSRQVH UDWH RI WKH /9, FRQGLWLRQ SUHYDLOHG
7DEOH 7KH VHTXHQFH RI FRQGLWLRQV DQG WKH QXPEHU RI VHVVLRQV FRQGXFWHG XQGHU HDFKf IRU ERWK VXEMHFWV LQ ([SHULPHQW XQGHU YDULDEOH LQWHUYDO 9,f OLQHDU YDULDEOH LQWHUYDO /9,f QHJDWLYH VORSH 16f DQG \RNHG /9, VFKHGXOHV RI SRLQW GHOLYHU\ 6XEMHFW 6 6 9, f 9, f /9, f /9, f 16 f 16 f 9, f 9, f /9, f 9, /+ V f 16 f 9, f 9, f /9, f /9, f
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f DQG ERWWRP SDQHOf $V LQ )LJXUH IURP ([SHULPHQW WKHVH SRLQWV KDYH EHHQ VXSHULPSRVHG XSRQ WKH IHHGEDFN IXQFWLRQV IRU WKH UHVSHFWLYH VFKHGXOHV 'XH WR WKH UHYHUVDO GHVLJQ HPSOR\HG LQ ([SHULPHQW WKH VXEMHFWV UHFHLYHG PXOWLSOH H[SRVXUHV WR WKH 9, DQG /9, FRQGLWLRQV )RU FODULW\ GDWD IURP VLPLODU FRQGLWLRQV VKDUH WKH VDPH V\PERO DQG WKH SRLQWV KDYH EHHQ
)LJXUH 0HDQ UDWH RI SRLQW GHOLYHU\ YHUVXV WKH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV XQGHU DOO FRQGLWLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 7KH GDWD KDYH EHHQ VXSHULPSRVHG XSRQ WKH PRODU IHHGEDFN IXQFWLRQV IRU WKH UHVSHFWLYH VFKHGXOHV 'DWD IURP VLPLODU FRQGLWLRQV VKDUH WKH VDPH V\PERO DQG WKH SRLQWV KDYH EHHQ QXPEHUHG DFFRUGLQJ WR WKH VHTXHQFH LQ ZKLFK ZHUH LPSOHPHQWHG 6HH 7DEOH DQG WKH WH[W IRU GHWDLOVf 1RWH LQGLYLGXDOO\ VFDOHG D[HV
6 2 /B 4 R &/ 6 L 16 L / 7 $ 9, V /9, V 16 )5 Q fÂ§2fÂ§ <2.( < 9, V /+ L ,c, QVHV
QXPEHUHG DFFRUGLQJ WR WKH VHTXHQFH LQ ZKLFK ZHUH LPSOHPHQWHG VHH 7DEOH f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t f WKDQ LQ WKH HQVXLQJ /9, FRQGLWLRQV 3RLQWV t UHVSHFWLYHO\f 7KH VWHDG\VWDWH UDWHV RI UHVSRQGLQJ DQG SRLQW GHOLYHU\ LQ WKH 16 FRQGLWLRQV 3RLQWV t f ZHUH ORZHU WKDQ WKRVH RI WKH LPPHGLDWHO\ SUHFHGLQJ /9, FRQGLWLRQV 3RLQWV t f LQGLFDWLQJ VHQVLWLYLW\ WR WKH QHJDWLYH VORSH FRQWLQJHQFLHV 7KH VWHDG\VWDWH UHVSRQVH UDWHV LQ WKH \RNHG /9, FRQGLWLRQ 3RLQW f ZHUH VXEVWDQWLDOO\ ORZHU WKDQ WKH VWHDG\VWDWH UDWHV IURP WKH SUHFHGLQJ /9, FRQGLWLRQ 3RLQW f DQG VOLJKWO\ ORZHU WKDQ WKH QHJDWLYH VORSH FRQGLWLRQ LQ ZKLFK WKH LQWHUSRLQW LQWHUYDOV ZHUH REWDLQHG 3RLQW f )RU 6XEMHFW ERWWRP SDQHOf VWHDG\VWDWH UDWHV RI UHVSRQGLQJ ZHUH KLJKHU LQ WKH ILUVW 9, FRQGLWLRQ 3RLQW f WKDQ LQ WKH VXEVHTXHQW /9, FRQGLWLRQ 3RLQW f ,Q WKH QHJDWLYHVORSH FRQGLWLRQ 3RLQW f VWHDG\VWDWH UDWHV RI UHVSRQGLQJ DQG SRLQW GHOLYHU\ GHFUHDVHG 8SRQ UHWXUQ WR WKH 9, FRQGLWLRQ 3RLQW f VWHDG\VWDWH UDWHV RI UHVSRQGLQJ LQFUHDVHG RQO\ VOLJKWO\ 7R UHFWLI\ WKLV D OLPLWHGKROG RQ WKH DYDLODELOLW\ RI SRLQWV VFKHGXOHG E\ WKH 9, V VFKHGXOH ZDV LPSOHPHQWHG $OWKRXJK UHVSRQVH UDWHV LQFUHDVHG
ZLWKLQ WKLV FRQGLWLRQ 3RLQW 6HH DOVR )LJXUH EHORZf WKH UDWHV GHFUHDVHG DJDLQ RQFH WKH OLPLWHG KROG ZDV GLVFRQWLQXHG 3RLQW f 6WHDG\VWDWH UHVSRQVH UDWHV GHFUHDVHG VOLJKWO\ LQ WKH VXEVHTXHQW /9, FRQGLWLRQ 3RLQW f $V ZLWK 6XEMHFW WKH VWHDG\VWDWH UHVSRQVH UDWHV LQ WKH \RNHG /9, FRQGLWLRQ 3RLQW f ZHUH ORZHU WKDQ WKH VWHDG\VWDWH UDWHV IURP WKH SUHFHGLQJ /9, FRQGLWLRQ 3RLQW f DQG IURP WKH QHJDWLYH VORSH FRQGLWLRQ LQ ZKLFK WKH LQWHUSRLQW LQWHUYDOV ZHUH REWDLQHG 3RLQW f 6HVVLRQE\6HVVLRQ 3DWWHUQV )LJXUH VKRZV WKH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV IRU ERWK VXEMHFWV :KHUHDV RQO\ VWHDG\VWDWH GDWD ZHUH SRUWUD\HG LQ )LJXUH WKLV ILJXUH VKRZV WUDQVLWLRQ GDWD DQG WUHQGV LQ UHVSRQVH DQG UHLQIRUFHPHQW UDWHV ZLWKLQ HDFK FRQGLWLRQ )RU 6XEMHFW WRS SDQHOf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
)LJXUH 7KH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 1RWH LQGLYLGXDOO\ VFDOHG D[HV
6HVVLRQV 5HVSRQVHV SHU 0LQXWH S R fÂ§ r Nf UR R /Q /Q /Q 3RLQWV SHU 0LQXWH 2 f 5HVSRQVHV 3RLQWV 6HVVLRQV 5HVSRQVHV SHU 0LQXWH .! WQ n2 R HQ R FQ 2 3RLQWV SHU 0LQXWH 2Q A
UHVSRQGLQJ DQG SRLQW GHOLYHU\ LQFUHDVHG WR OHYHOV FRPSDUDEOH WR WKRVH RI WKH SUHYLRXV H[SRVXUHV LQGLFDWLQJ D UHFRYHU\ RI EDVHOLQH SHUIRUPDQFH :KHQ WKH \RNHG /9, FRQGLWLRQ ZDV LPSOHPHQWHG UHVSRQVH UDWHV ZHUH VOLJKWO\ ORZHU WKDQ WKH UDWHV LQ WKH SUHFHGLQJ /9, FRQGLWLRQ DQG DSSUR[LPDWHO\ HTXDO WR WKH UDWHV LQ WKH 16 FRQGLWLRQ E\ WKH FRQFOXVLRQ 7KH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV XQGHU WKH \RNHG /9, FRQGLWLRQ ZDV b RI WKH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV RI WKH LPPHGLDWHO\ SUHFHGLQJ /9, FRQGLWLRQ ZKHUHDV WKH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV RI WKH 16 FRQGLWLRQ ZDV b RI WKH LPPHGLDWHO\ SUHFHGLQJ /9, FRQGLWLRQ 1RWH WKH KLJK GHJUHH RI FRUUHVSRQGHQFH EHWZHHQ WKH UDWHV RI SRLQW GHOLYHUfn LQ WKH 16 DQG \RNHG /9, FRQGLWLRQV ZKLFK LV LQ DFFRUG ZLWK WKH SURJUDPPHG FRQWLQJHQFLHV 7KH RQO\ GHYLDWLRQV IURP H[DFW DFFRUGDQFH RFFXUUHG LQ 6HVVLRQV DQG 7KH VWRUH ZDV SRVLWLYH DW WKH FRQFOXVLRQ RI 6HVVLRQ ZLWK D SRLQW FDUULHG RYHU WR 6HVVLRQ $V D UHVXOW WKH RYHUDOO UDWH RI SRLQW GHOLYHU\ ZDV VOLJKWO\ ORZHU LQ 6HVVLRQ WKDQ LQ 6HVVLRQ WKH VHVVLRQ WR ZKLFK LW ZDV \RNHGf DQG VOLJKWO\ KLJKHU LQ 6HVVLRQ WKDQ LQ 6HVVLRQ 7KH ERWWRP SDQHO RI )LJXUH VKRZV WKH PHDQ UHVSRQVH UDWHV ILOOHG FLUFOHVf DQG PHDQ UDWHV RI SRLQW GHOLYHU\ RSHQ FLUFOHVf DFURVV VHVVLRQV IRU DOO FRQGLWLRQV IRU 6XEMHFW (DUO\ LQ WKH LQLWLDO H[SRVXUH WR WKH 9, FRQGLWLRQ UHVSRQVH UDWHV ZHUH ORZ EXW WKHQ LQFUHDVHG DFURVV VHVVLRQV 5DWHV RI SRLQW GHOLYHU\ DOVR LQFUHDVHG JUDGXDOO\ DFURVV WKH FRQGLWLRQ ,Q WKH VXEVHTXHQW /9, FRQGLWLRQ KRZHYHU UHVSRQVH UDWHV GHFUHDVHG FRQVLGHUDEO\ ZKLOH UDWHV RI SRLQW GHOLYHU\ LQFUHDVHG VOLJKWO\
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
WKH UDWHV IURP WKH FRUUHVSRQGLQJ QHJDWLYH VORSH VHVVLRQV 7KH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV XQGHU WKH \RNHG /9, FRQGLWLRQ ZDV b RI WKH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV RI WKH LPPHGLDWHO\ SUHFHGLQJ /9, FRQGLWLRQ ZKHUHDV WKH PHDQ UHVSRQVH UDWH IURP WKH ODVW VHVVLRQV RI WKH 16 FRQGLWLRQ ZDV b RI WKH LPPHGLDWHO\ SUHFHGLQJ /9, FRQGLWLRQ :LWKLQ6HVVLRQ 3DWWHUQV 6HFWLRQV RI UHSUHVHQWDWLYH FXPXODWLYH UHFRUGV IURP VHVVLRQV FRQGXFWHG XQGHU 16 DQG \RNHG /9, FRQGLWLRQV DUH VKRZQ LQ )LJXUH IRU 6XEMHFW WRS SDQHOVf DQG ERWWRP SDQHOVf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
)LJXUH 3RUWLRQV RI UHSUHVHQWDWLYH FXPXODWLYH UHVSRQVH UHFRUGV IURP VHVVLRQV FRQGXFWHG XQGHU 16 DQG \RNHG /9, FRQGLWLRQV IRU 6XEMHFWV DQG (DFK SDQHO WKH VHJPHQW RI WKH FXPXODWLYH UHFRUG IURP WKH VHFRQG PLQ EORFN RI WKH VHVVLRQ
&XPXODWLYH 5HVSRQVHV &XPXODWLYH 5HVSRQVHV 16 6HVVLRQ 6 <2.(' 6HVVLRQ 16 <2.(' 6HVVLRQ 6HVVLRQ 9, LQ XWHV
,57V 3UHFHGLQJ 3RLQW 'HOLYHU\ 6FDWWHU SORWV RI WKH ,57V SUHFHGLQJ SRLQW GHOLYHU\ DFURVV VHVVLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf DUH VKRZQ LQ )LJXUH 7KH UHOHYDQW FRPSDULVRQV LQ WKLV ILJXUH DUH EHWZHHQ WKH GLVWULEXWLRQV IURP WKH \RNHG /9, FRQGLWLRQV DQG WKH GLVWULEXWLRQV IURP WKH QHJDWLYH VORSH FRQGLWLRQV LQ ZKLFK LQWHUSRLQW LQWHUYDOV ZHUH REWDLQHG 16 DQG 16 IRU 6XEMHFWV DQG UHVSHFWLYHO\f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
)LJXUH 6FDWWHU SORWV RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ DFURVV VHVVLRQV IRU 6XEMHFWV WRS SDQHOf DQG ERWWRP SDQHOf 1RWH LQGLYLGXDOO\ VFDOHG DQG ORJDULWKPLF D[HV
6HVVLRQV ,57 3UHFHGLQJ 3RLQW 'HOLYHU\ 6HFf ,57 3UHFHGLQJ 3RLQW 'HOLYHU\
'LVFXVVLRQ 7KH UHVXOWV IURP WKH QHJDWLYH VORSH FRQGLWLRQV RI ([SHULPHQW UHSOLFDWH WKRVH RI ([SHULPHQW 5HVSRQVH UDWHV ZHUH ORZHU LQ WKH QHJDWLYH VORSH FRQGLWLRQV WKDQ LQ WKH LQLWLDO 9, EDVHOLQH FRQGLWLRQV EXW ZHUH QRW ORZ HQRXJK WR PD[LPL]H RYHUDOO UDWH RI SRLQW GHOLYHU\ 8QOLNH UDWV 5HHG t 6FKDFKWPDQ f DQG SLJHRQV 9DXJKDQ f KRZHYHU WKH UHVSRQVH UDWHV RI ERWK VXEMHFWV ZHUH VOLJKWO\ ORZHU LQ WKH \RNHG /9, FRQGLWLRQ WKDQ LQ WKH FRUUHVSRQGLQJ QHJDWLYHVORSH FRQGLWLRQ VHH )LJXUHV t f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f DQG ELPRGDO GLVWULEXWLRQV RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ ZHUH REWDLQHG IRU 6XEMHFW RQO\ VHH )LJXUH f ,I WKHVH FKDUDFWHULVWLFV RI SHUIRUPDQFH DUH W\SLFDO RI WHUPLQDO SHUIRUPDQFH RQ QHJDWLYH VORSH VFKHGXOHV WKHQ SHUKDSV WKH VXEMHFWV LQ ([SHULPHQW GLG QRW UHFHLYH VXIILFLHQW H[SRVXUH WR WKH QHJDWLYH VORSH FRQWLQJHQFLHV 7KH UHYHUVDO GHVLJQ HPSOR\HG LQ ([SHULPHQW PD\ DOVR KDYH SURGXFHG WURXEOHVRPH KLVWRU\ HIIHFWV WKDW FRPSOLFDWH LQWHUSUHWDWLRQ RI WKH UHVXOWV )RU 6XEMHFW UHVSRQVH
UDWHV GHFUHDVHG LQ WKH ILUVW H[SRVXUH WR WKH /9, FRQGLWLRQ DQG IROORZLQJ H[SRVXUH WR WKH QHJDWLYH VORSH FRQWLQJHQF\ LW ZDV QRW SRVVLEOH WR UHFRYHU EDVHOLQH OHYHOV RI SHUIRUPDQFH ([SRVXUH WR WKHVH FRQGLWLRQV SRVVLEO\ DWWHQXDWHG VHQVLWLYLW\ WR DGGLWLRQDO PDQLSXODWLRQV RI WKH FRQWLQJHQFLHV 7KHVH UHVXOWV PD\ QRW EH WRR VXUSULVLQJ JLYHQ WKH SUHYDOHQFH RI KXPDQ EHKDYLRUDO SHUVLVWHQFH XQGHU ODERUDWRU\ FRQGLWLRQV :HLQHU f $V :HLQHU f KDV QRWHG KXPDQ UHVSRQVH SDWWHUQV WHQG WR SHUVLVW XQOHVV DGYHUVH UHLQIRUFHPHQW FRQVHTXHQFHV LQGXFH FKDQJHf S f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
6XEMHFW ZURWH ,W VHHPHG WR PH SRLQWV ZHUH RQO\ DYDLODEOH LI \RX SUHVVHG WKH UHG EXWWRQ %XW ZLWKLQ WKH WHVWV LW VHHPHG WR PH WKDW LI \RX SUHVVHG WKH UHG EXWWRQ WRR RIWHQ SRLQWV ZHUH PRUH HOXVLYH ZKLFK OHG PH WR D VWUDWHJ\ RI UHVWUDLQW LQ KRZ RIWHQ SXVKHG WKH EXWWRQ $OWKRXJK WKH RULJLQ DQG FDXVDO VWDWXV RI WKLV YHUEDO UHSRUW DUH GXELRXV LW LV FRQVLVWHQW ZLWK KLV QRQYHUEDO SHUIRUPDQFH DQG VXJJHVWV WKDW DW OHDVW VRPH DVSHFW RI WKH VXEMHFWnV EHKDYLRU ZDV VHQVLWLYH WR WKH QHJDWLYHVORSH FRQWLQJHQFLHV 7KH EHKDYLRU SURGXFHG E\ WKH QHJDWLYH VORSH FRQGLWLRQ YHUEDO DQG QRQYHUEDOf PD\ KDYH DOVR EHHQ RFFDVLRQHG LQ WKH \RNHG /9, FRQGLWLRQ E\ WKH VXGGHQ GHFUHDVH LQ UDWH RI SRLQW GHOLYHU\ ,W LV WKHQ SRVVLEOH WKDW WKH FRQWLQJHQFLHV RI WKH \RNHG /9, FRQGLWLRQ PD\ KDYH VXVWDLQHG WKHVH SDWWHUQV RI EHKDYLRU 7KDW LV 6XEMHFW PD\ KDYH EHHQ VHQVLWLYH WR SUHVHQFH RI WKH VXEWUDFWLRQ FRQWLQJHQF\ LQ WKH QHJDWLYH VORSH FRQGLWLRQ EXW QRW VHQVLWLYH WR WKH DEVHQFH RI WKH FRQWLQJHQF\ LQ WKH \RNHG /9, FRQGLWLRQ 3HUIRUPDQFH LQ WKH \RNHG /9, FRQGLWLRQ PD\ KDYH EHHQ DQ LQVWDQFH RI ZKDW 6KLPRII &DWDQLD DQG 0DWKHZV f UHIHUUHG WR DV SVHXGRVHQVLWLYLW\ WR WKH FRQWLQJHQFLHVUXOHJRYHPHG UHVSRQVH SDWWHUQV LQ DFFRUG ZLWK WKH FRQWLQJHQFLHV ,Q WKH SUHVHQW FDVH WKH UHVSRQVH UDWHV RQ WKH \RNHG /9, PD\ KDYH EHHQ SURGXFHG GLUHFWO\ E\ WKH UDWH RI SRLQW GHOLYHU\ $OWHUQDWLYHO\ WKH\ PD\ KDYH EHHQ LQIOXHQFHG E\ YHUEDO GHVFULSWLRQV RI WKH FRQWLQJHQFLHV JHQHUDWHG XQGHU WKH QHJDWLYH VORSH FRQGLWLRQV DQG RFFDVLRQHG LQ WKH \RNHG /9, FRQGLWLRQ ,Q VXP WKH UHVXOWV RI ([SHULPHQW DUH LQFRQFOXVLYH ZLWK UHVSHFW WR WKH TXHVWLRQ RI ZKHWKHU KXPDQ EHKDYLRU LV VHQVLWLYH WR WKH SUHVHQFH RU DEVHQFH RI WKH VXEWUDFWLRQ FRQWLQJHQF\ $W OHDVW VRPH RI WKH SUREOHPV PD\ KDYH EHHQ FLUFXPYHQWHG LI WKH LQWHUSRLQW LQWHUYDOV ZHUH \RNHG DFURVV FRPSRQHQWV RI D PXOWLSOH VFKHGXOH UDWKHU WKDQ
DFURVV H[SHULPHQWDO FRQGLWLRQV 7KH DGGLWLRQ RI GLVFULPLQDWLYH VWLPXOL IRU HDFK RI WKH GLIIHUHQW FRQGLWLRQV PD\ KDYH KHOSHG IRVWHU VHQVLWLYLW\ WR WKH GLIIHUHQFHV EHWZHHQ WKH \RNHG DQG WKH QHJDWLYH VORSH FRQGLWLRQV DQG PLQLPL]HG WKH SHUVLVWHQFH RI SUHHVWDEOLVKHG UHVSRQVH SDWWHUQV
*(1(5$/ ',6&866,21 7KH RYHUDOO SDWWHUQ RI UHVXOWV IURP ERWK H[SHULPHQWV VXJJHVWV WKDW KXPDQ EHKDYLRU LV VHQVLWLYH WR WKH FRQWLQJHQFLHV RI QHJDWLYHVORSH VFKHGXOHV 7KLV ZDV HYLGHQFHG SULPDULO\ E\ WKH GHFUHDVHV LQ RYHUDOO UDWHV RI UHVSRQGLQJ DFURVV QHJDWLYHVORSH FRQGLWLRQV LQ ([SHULPHQW 6HH )LJXUH f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nV f H[SHULPHQW KRZHYHU $FURVV VXEMHFWV WKH UHVSRQVH UDWHV WHQGHG WR IDOO FORVHU WR WKH DSH[HV RI WKH IHHGEDFN IXQFWLRQ WKDQ WKRVH RI WKH SLJHRQV 0RUHRYHU UHLQIRUFHPHQW UDWHV RFFDVLRQDOO\ LQFUHDVHG IRU KXPDQV DV WKH VXEWUDFWLRQ UDWLRV EHFDPH PRUH VWULQJHQW ZKHUHDV WKH SLJHRQV WHQGHG WR SHUIRUP SRRUO\ DW WKH ORZHU VXEWUDFWLRQ UDWLRV 7KHVH UHVXOWV FRPSOHPHQW WKH UHVXOWV RI H[SHULPHQWV H[DPLQLQJ SHUIRUPDQFH RQ FRQFXUUHQW UHVSRQVHEDVHG DQG WLPHEDVHG VFKHGXOHV 6LOEHUEHUJ 7KRPDV t %HUHQG]HQ 6DYDVWDQR t )DQWLQR f 2Q WKHVH SURFHGXUHV UHLQIRUFHPHQW UDWH LV PD[LPL]HG E\ UHVSRQGLQJ PRVWO\ RQ WKH UHVSRQVHEDVHG DOWHUQDWLYH WKH WLPHEDVHG VFKHGXOH VKRXOG EH VDPSOHG RQO\ EULHIO\ DQG RFFDVLRQDOO\ LQ RUGHU WR FROOHFW UHLQIRUFHUV DV WKH\ DUH VHW XS %RWK SLJHRQV +H\PDQ t +HUPVWHLQ f DQG KXPDQV 6LOEHUEHUJ HW DO 6DYDVWDQR t )DQWLQR f DOORFDWH WRR PXFK WLPH WR WKH WLPHEDVHG DOWHUQDWLYH DQG RYHUDOO UHLQIRUFHPHQW UDWH LV QRW PD[LPL]HG +XPDQV KRZHYHU WHQG WR VKRZ D JUHDWHU ELDV IRU WKH UDWLR DOWHUQDWLYH WKDQ GR SLJHRQV 7KH SUHVHQW UHVXOWV DUH DW RGGV ZLWK WKRVH RI SULRU H[SHULPHQWV H[DPLQLQJ VHOI FRQWURO /RJXH HW DO f DQG FKRLFH LQ VLWXDWLRQV RI GLPLQLVKLQJ UHWXUQV +DFNHQEHUJ
t $[WHOO -DFREV t +DFNHQEHUJ :DQFKLVHQ HW DO f LQ KXPDQV ZKLFK DUH JHQHUDOO\ FRQVLVWHQW ZLWK DFFRXQWV EDVHG XSRQ WKH PD[LPL]DWLRQ RI RYHUDOO UHLQIRUFHPHQW GHQVLW\ ,QGHHG GHYLDWLRQV IURP PD[LPL]DWLRQ KDYH EHFRPH WKH H[FHSWLRQ UDWKHU WKDQ WKH UXOH LQ ODERUDWRU\ H[SHULPHQWV ZLWK DGXOW KXPDQ VXEMHFWV ,W PD\ EH GHVLUDEOH WR VHDUFK WKHVH IHZ H[FHSWLRQDO FDVHV IRU FRPPRQ IHDWXUHV 7KLV WDVN LV FRPSOLFDWHG KRZHYHU E\ VLJQLILFDQW PHWKRGRORJLFDO GLIIHUHQFHV DFURVV H[SHULPHQWV ,Q DQ H[SHULPHQW E\ :DVVHUPDQ DQG 1HXQDEHU f IRU H[DPSOH SUHVVLQJ D WHOHJUDSK NH\ ZDV PDLQWDLQHG LQ VWXGHQW YROXQWHHUV E\ FRQWLJXRXV SRLQW SUHVHQWDWLRQV HYHQ ZKHQ SUHVVLQJ DOVR GHFUHDVHG WKH RYHUDOO UDWH RI SRLQW GHOLYHU\ 7KHVH UHVXOWV VXJJHVW WKDW KXPDQ EHKDYLRU FDQ XQGHU FHUWDLQ FLUFXPVWDQFHV EH PDLQWDLQHG E\ VKRUWWHUP FRQVHTXHQFHV DW WKH H[SHQVH RI ORQJWHUP RXWFRPHV ,W LV LPSRUWDQW WR QRWH KRZHYHU WKDW H[SRVXUH WR WKH FRQWLQJHQFLHV ZDV EULHI VHVVLRQf DQG WKH SRLQWV ZHUH QRW H[FKDQJHDEOH IRU RWKHU UHLQIRUFHUV LH WKH\ ZHUH PHUHO\ fLQVWUXFWHG UHLQIRUFHUVf 7KH HIIHFWV FDQ WKHUHIRUH EH TXHVWLRQHG RQ WKH JURXQGV WKDW WKH VXEMHFWV GLG QRW KDYH VXIILFLHQW H[SRVXUH WR WKH FRQWLQJHQFLHV WR FRQWDFW WKH ORQJWHUP FRQVHTXHQFHV DQG WKDW WKH SRLQW SUHVHQWDWLRQV ZHUH ZHDN UHLQIRUFHUV LI LQGHHG WKH\ ZHUH UHLQIRUFHUV DW DOOf ,PSOLFDWLRQV IRU 0DWFKLQJ $FFRXQWV 0DWFKLQJEDVHG LQWHUSUHWDWLRQV GR QRW DFFRXQW ZHOO IRU WKH VWHDG\VWDWH GDWD HLWKHU $V GLVFXVVHG DERYH WKH VWHDG\VWDWH GDWD IURP ([SHULPHQW SRUWUD\HG LQ )LJXUH GR QRW FRQIRUP WR (TXDWLRQ 7KLV LV LQ FRQWUDVW ZLWK WKH UHVXOWV RI VHYHUDO RWKHU VWXGLHV H[DPLQLQJ KXPDQ SHUIRUPDQFH RQ 9, VFKHGXOHV RI UHLQIRUFHPHQW SUHVHQWHG HLWKHU DORQH RU FRQFXUUHQWO\ %DXP %UDGVKDZ 6]DEDGL t %HYDQ
0F'RZHOO t :RRG 6KURHGHU t +ROODQG EXW VHH +RPH t /RZH 6FKPLWW f ,Q WZR VWXGLHV E\ %UDGVKDZ DQG FROOHDJXHV f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r1 UHVSRQVH RQ DYHUDJH SURGXFHG SRLQW ORVV VLJQDOHG E\ D PV LOOXPLQDWLRQ RI D UHG ODPS DQG WKH VXEWUDFWLRQ RI RQH SRLQW IURP WKH FRXQWHU $OWKRXJK WKHUH DUH PDQ\ GLIIHUHQFHV EHWZHHQ WKLV SURFHGXUH DQG WKDW RI WKH QHJDWLYH VORSH FRQGLWLRQV RI WKH SUHVHQW VWXG\ ERWK HVWDEOLVKHG DQ LQYHUVH UHODWLRQVKLS EHWZHHQ RYHUDOO UHVSRQVH UDWH DQG RYHUDOO UHLQIRUFHPHQW UDWH 7KH QHW UHVXOW ZDV WKH VDPH KLJKHU UDWHV RI UHVSRQGLQJ UHVXOWHG LQ ORZHU UDWHV RI UHLQIRUFHPHQW RYHUDOO 7KH PDMRU GLIIHUHQFH EHWZHHQ WKH WZR SURFHGXUHV OLHV LQ WKH GLVFULPLQDELOLW\ RI SRLQW ORVV ,Q WKH %UDGVKDZ HW DO VWXGLHV f SRLQW ORVV ZDV VLJQDOHG ZKHUHDV WKH SRLQW RPLVVLRQV LQ WKH SUHVHQW VWXG\ ZHUH XQVLJQDOHG $V LQ WKH QHJDWLYHVORSH FRQGLWLRQV RI WKH SUHVHQW VWXG\ UHVSRQVH UDWHV LQ WKH SXQLVKPHQW FRQGLWLRQV RI WKH %UDGVKDZ HW DO VWXGLHV f ZHUH ORZHU WKDQ
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f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f SHUIRUPHG D V\VWHPDWLF UHSOLFDWLRQ RI WKH %UDGVKDZ HW DO f VWXG\ DQG SURGXFHG HYLGHQFH VXJJHVWLQJ WKDW WKH UHVXOWV FRXOG EH H[SODLQHG RQ WKH EDVLV RI LPSOLFLW RUGLQDO VWLPXOL ,Q D VHULHV RI H[SHULPHQWV WKH\ IRXQG WKDW PDWFKLQJ ZDV
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f RIIHUHG D JHQHUDO GHVFULSWLRQ WKH UHODWLRQVKLS EHWZHHQ UDWH RI SRLQW GHOLYHU\ DQG UDWH RI UHVSRQGLQJ QRQH RI WKH VXEMHFWV ZHUH DEOH WR VSHFLI\ D PD[LPL]DWLRQ VWUDWHJ\ DV WKH VXEMHFWV LQ WKH VHOIFRQWURO H[SHULPHQWV FRXOG /RJXH HW DO f 7KH DSHULRGLFLW\ RI SRLQW GHOLYHU\ DUUDQJHG E\ WKH SUHVHQW SURFHGXUHV PD\ KDYH EHHQ VXIILFLHQWO\ VXEWOH WR SUHFOXGH DFFXUDWH GHVFULSWLRQV WKHUHE\ HOXGLQJ WKH LQIOXHQFH RI YHUEDO PHGLDWLRQ WR D VLJQLILFDQW GHJUHH %HFDXVH WKH SUHVHQW VWXG\ ZDV QRW GHVLJQHG WR FRQWURO YHUEDO EHKDYLRU QRU WR PHDVXUH LWV UHODWLRQ WR LQ QRQYHUEDO SHUIRUPDQFH LW LV QRW SRVVLEOH WR UXOH RXW YHUEDO PHGLDWLRQ 7R GR VR ZRXOG UHTXLUH UHFRUGLQJ RQJRLQJ YHUEDO EHKDYLRU DQG LWV UHODWLRQ WR WKH
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f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
QHJDWLYH VORSH VFKHGXOH SHUIRUPDQFH SURYLGH GDWD RI WKLV VRUW ,Q WKH SUHVHQW VWXG\ WKH ,57V SUHFHGLQJ SRLQW GHOLYHULHV ZHUH FROOHFWHG DFURVV DOO FRQGLWLRQV IRU RI VXEMHFWV 7KH GLVWULEXWLRQ RI ,57V SUHFHGLQJ SRLQW GHOLYHU\ EHFDPH ELPRGDO DQG WKH UDQJH LQFUHDVHG XQGHU WKH QHJDWLYH VORSH FRQGLWLRQV 6HH )LJXUHV t f LQ RI WKHVH VXEMHFWV 7KHVH GDWD DUH FRQVLVWHQW ZLWK 5HHG DQG 6FKDFKWPDQnV f K\SRWKHVLV WKDW UHODWLYHO\ ORQJHU ,57V DUH GLIIHUHQWLDOO\ UHLQIRUFHG XQGHU QHJDWLYH VORSH VFKHGXOHV 7KH ELPRGDO GLVWULEXWLRQV REWDLQHG LQ WKH SUHVHQW VWXG\ UHYHDO VRPHWKLQJ WKDW 5HHG DQG 6FKDFKWPDQn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f WKH\ DUH VWLOO XVHIXO IRU DVVHVVLQJ VFKHGXOH FRQWURO E\ SRUWUD\LQJ VFKHGXOHW\SLFDO PRPHQWWRPRPHQW IOXFWXDWLRQV LQ UHVSRQGLQJ )HUVWHU t 6NLQQHU
=HLOHU f ,Q DOO VWXGLHV HPSOR\LQJ QHJDWLYHVORSH VFKHGXOHV WR GDWH (WWLQJHU HW DO 5HHG t 6FKDFKWPDQ 9DXJKDQ t 0LOOHU f KRZHYHU WKHUH KDV EHHQ RQO\ RQH FXPXODWLYH UHFRUG SXEOLVKHG (WWLQJHU HW DO f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f VKRZLQJ WKLV SDWWHUQ W\SLFDO 9, SDWWHUQLQJ ZDV HVWDEOLVKHG LQ WKH LQLWLDO 9, FRQGLWLRQ ZDV PDLQWDLQHG WKURXJKRXW WKH /9, FRQGLWLRQ WKHQ WUDQVLWLRQHG WR WKH SDXVH DQG UXQ SDWWHUQ DFURVV QHJDWLYHVORSH FRQGLWLRQV 7KH
GHYHORSPHQW RI WKLV SDWWHUQ SURYLGHV IXUWKHU HYLGHQFH WKDW WKH VXEMHFWVn EHKDYLRU ZDV VHQVLWLYH WR WKH VXEWUDFWLRQ FRQWLQJHQF\ ,I UHVSRQGLQJ ZDV VLPSO\ WUDFNLQJ FKDQJHV LQ RYHUDOO UDWH RI SRLQW GHOLYHU\ WKHUH LV QR UHDVRQ WR VXSSRVH WKDW UHVSRQVH SDWWHUQV ZRXOG GLIIHU DFURVV WKH 9, DQG QHJDWLYHVORSH FRQGLWLRQV ,QGHHG RQH PLJKW H[SHFW GLIIHUHQFHV WR EH VHHQ SULPDULO\ LQ WKH RYHUDOO UDWHV QRW LQ WKH ORFDO SDWWHUQLQJ 2I WKH VXEMHFWV 6XEMHFWV DQG f WKDW GLG QRW VKRZ WKH SDXVHUXQf SDWWHUQ UREXVWO\ DOO UHVSRQGHG DW PXFK ORZHU UDWHV LQ WKH EDVHOLQH 9, DQG /9, FRQGLWLRQV WKDQ WKH RWKHU VXEMHFWV ZLWK WZR 6XEMHFWV DQG f VKRZLQJ VHQVLWLYLW\ WR WKH 9, /9, GLIIHUHQFH 0RUHRYHU WKHVH VXEMHFWV JHQHUDOO\ UHFHLYHG OHVV H[SRVXUH RYHUDOO WR WKH QHJDWLYH VORSH VFKHGXOHV WKDQ VRPH RI WKH RWKHU VXEMHFWV ,W LV SRVVLEOH WKDW WKHVH VXEMHFWV GLG QRW UHFHLYH VXIILFLHQW H[SRVXUH WR WKH QHJDWLYH VORSH VFKHGXOHV IRU WKH fSDXVHUXQnf SDWWHUQ WR GHYHORS ,W LV DOVR SRVVLEOH WKDW WKH fSDXVHUXQf SDWWHUQ RI UHVSRQGLQJ GHSHQGV RQ WKH EDVHOLQH UHVSRQVH UDWH 7KH UHVSRQVH SDWWHUQV RI KXPDQV RQ IL[HG LQWHUYDO VFKHGXOHV IRU H[DPSOH JHQHUDOO\ IDOO LQWR RQH RI WZR FDWHJRULHV ORZ UDWH FKDUDFWHUL]HG E\ ORQJ SDXVHV ZLWK D IHZ UHVSRQVHV DW WKH HQG RI WKH LQWHUYDO RU KLJK UDWH FKDUDFWHUL]HG E\ LQIUHTXHQW SDXVLQJ DQG YLUWXDO LQVHQVLWLYLW\ WR VFKHGXOH FKDQJHV /DWLHV t :HLVV /LSSPDQ t 0H\HU f :HLQHU f KDV GHPRQVWUDWHG WKDW D SULRU H[SHULPHQWDO KLVWRU\ FDQ LQIOXHQFH WKH SDWWHUQ RI ), UHVSRQGLQJ 6XEMHFWV LQLWLDOO\ H[SRVHG WR DQ ,57 W V VFKHGXOH WHQGHG WR UHVSRQG DW D ORZ UDWH ZKHQ VXEVHTXHQWO\ H[SRVHG WR D VHULHV RI ), VFKHGXOHV ZKHUHDV WKRVH LQLWLDOO\ H[SRVHG WR D IL[HGUDWLR VFKHGXOH WHQGHG WR UHVSRQG DW D KLJK FRQVWDQW UDWH XQGHU WKH ), VFKHGXOHV ,W LV SRVVLEOH WKDW WKH VXEMHFWVn SUHH[SHULPHQWDO KLVWRULHV DQGRU
WKHLU ORZUDWH KLVWRULHV RQ WKH EDVHOLQH 9, DQG /9, FRQGLWLRQV PRGXODWHG VHQVLWLYLW\ WR WKH QHJDWLYHVORSH FRQGLWLRQV RI WKH SUHVHQW H[SHULPHQW &XPXODWLYH UHVSRQVH UHFRUGV PD\ EH D XVHIXO WRRO IRU WKH LGHQWLILFDWLRQ RI UHVSRQVH XQLWV FUHDWHG DQG PDLQWDLQHG XQGHU QHJDWLYH VORSH VFKHGXOHV 7KH fSDXVH DQG UXQf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f PD\ RQH GD\ EH EURXJKW WR EHDU RQ QHJDWLYH VORSH SHUIRUPDQFH DV ZHOO
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
5()(5(1&(6 $LQVOLH : f ,PSXOVH FRQWURO LQ SLJHRQV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU %DXP : 0 f 7KH FRUUHODWLRQEDVHG ODZ RI HIIHFW -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU %DXP : 0 f 7LPH DOORFDWLRQ LQ KXPDQ YLJLODQFH -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU %UDGVKDZ & 0 6]DEDGL ( t %HYDQ 3 f %HKDYLRU RI KXPDQV LQ YDULDEOH LQWHUYDO VFKHGXOHV RI UHLQIRUFHPHQW -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU %UDGVKDZ & 0 6]DEDGL ( t %HYDQ 3 f (IIHFW RI SXQLVKPHQW RQ KXPDQ YDULDEOHLQWHUYDO SHUIRUPDQFH -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU %UDGVKDZ & 0 6]DEDGL ( t %HYDQ 3 f 7KH HIIHFW RI SXQLVKPHQW RQ IUHH RSHUDQW FKRLFH LQ KXPDQV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU &DWDQLD $ & 0DWKHZV % $ t 6KLPRII ( f ,QVWUXFWHG YHUVXV VKDSHG KXPDQ YHUEDO EHKDYLRU ,QWHUDFWLRQV ZLWK QRQYHUEDO UHVSRQGLQJ -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU &DWDQLD $ & t 5H\QROGV 6 f $ TXDQWLWDWLYH DQDO\VLV RI WKH UHVSRQGLQJ PDLQWDLQHG E\ LQWHUYDO VFKHGXOHV RI UHLQIRUFHPHQW -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU &HUXWWL 7 f 'LVFULPLQDWLYH WKHRU\ RI UXOHJRYHUQHG EHKDYLRU -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU B (WWLQJHU 5 + 5LHG $ t 6WDGGRQ ( 5 f 6HQVLWLYLW\ WR 0RODU )HHGEDFN )XQFWLRQV $ 7HVW RI 0RODU 2SWLPDOLW\ 7KHRU\ -RXUQDO RI ([SHULPHQWDO 3V\FKRORJ\ $QLPDO %HKDYLRU 3URFHVVHV )HUVWHU & % t 6NLQQHU %) f 6FKHGXOHV RI UHLQIRUFHPHQW 1HZ
)OHVKLHU 0 t +RIIPDQ + 6 f $ SURJUHVVLRQ IRU JHQHUDWLQJ YDULDEOHLQWHUYDO VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +DFNHQEHUJ 7 t $[WHOO 6 $ 0 f +XPDQVn FKRLFHV LQ VLWXDWLRQV RI WLPH EDVHG GLPLQLVKLQJ UHWXUQV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +DFNHQEHUJ 7 t +LQHOLQH 3 1 f &KRLFH LQ VLWXDWLRQV RI GLPLQLVKLQJ UHWXUQV ,PPHGLDWH YHUVXV GHOD\HG FRQVHTXHQFHV RI DFWLRQ -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +HUPVWHLQ 5 f 2Q WKH ODZ RI HIIHFW -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +H\PDQ 0 t +HUPVWHLQ 5 f 0RUH FRQFXUUHQW LQWHUYDOUDWLR VFKHGXOHV $ UHSOLFDWLRQ DQG UHYLHZ -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +LQHOLQH 3 1 t 6RGHW] ) f $SSHWLWLYH DQG DYHUVLYH VFKHGXOH SUHIHUHQFHV 6FKHGXOH WUDQVLWLRQV DV LQWHUYHQLQJ HYHQWV ,Q 0 / &RPPRQV ( 0D]XU $ 1HYLQ t + 5DFKOLQ (GVf 4XDQWLWDWLYH DQDO\VHV RI EHKDYLRU 9RO 7KH HIIHFW RI GHOD\ DQG LQWHUYHQLQJ HYHQWV RQ UHLQIRUFHPHQW YDOXH SS f +LOOVGDOH 1(UOEDXP +RGRV : t 7UXPEXOH + f 6WUDWHJLHV RI VFKHGXOH SUHIHUHQFH LQ FKLPSDQ]HHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU -B +RUQH 3 t /RZH & ) f 'HWHUPLQDQWV RI KXPDQ SHUIRUPDQFH RQ FRQFXUUHQW VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +RUQH 3 t /RZH & ) f 2Q WKH RULJLQV RI QDPLQJ DQG RWKHU V\PEROLF EHKDYLRU -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU +\WHQ & 0DGGHQ DQG )LHOG 3 f ([FKDQJH GHOD\V DQG LPSXOVLYH FKRLFH LQ DGXOW KXPDQV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU -DFNVRQ t +DFNHQEHUJ 7 f 7RNHQ UHLQIRUFHPHQW FKRLFH DQG VHOIFRQWURO LQ SLJHRQV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU
-DFREV ( $ t +DFNHQEHUJ 7 f +XPDQVn FKRLFHV LQ VLWXDWLRQV RI GLPLQLVKLQJ UHWXUQV (IIHFWV RI IL[HGLQWHUYDO GXUDWLRQ DQG SURJUHVVLYHLQWHUYDO VWHS VL]H -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU /RJXH $ : f 5HVHDUFK RQ VHOIFRQWURO $Q LQWHJUDWLQJ IUDPHZRUN %HKDYLRUDO t %UDLQ 6FLHQFHV /RJXH $: t .LQJ *5 f 6HOIFRQWURO DQG LPSXOVLYHQHVV LQ DGXOW KXPDQV ZKHQ IRRG LV WKH UHLQIRUFHU $SSHWLWH /RJXH $: 3HQD&RUUHDO 7( 5RGULJXH] 0/ t .DEHOD ( f 6HOIFRQWURO LQ DGXOW KXPDQV 9DULDWLRQ LQ SRVLWLYH UHLQIRUFHU DPRXQW DQG GHOD\ -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU .LOOHHQ 3 f 0DWKHPDWLFDO SULQFLSOHV RI UHLQIRUFHPHQW %HKDYLRUDO DQG %UDLQ 6FLHQFH 0D]XU ( t 9DXJKDQ : -U f 0RODU RSWLPL]DWLRQ YHUVXV GHOD\HG UHLQIRUFHPHQW DV H[SODQDWLRQV RI FKRLFH EHWZHHQ IL[HGUDWLR DQG SURJUHVVLYHUDWLR VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 0F'RZHOO -t :RRG +0 f &RQILUPDWLRQ RI OLQHDU V\VWHPV WKHRU\ SUHGLFWLRQ &KDQJHV LQ +HUUQVWHLQnV N DV D IXQFWLRQ RI FKDQJHV LQ UHLQIRUFHU PDJQLWXGH -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 1DYDULFN f 1HJDWLYH UHLQIRUFHPHQW DQG FKRLFH LQ KXPDQV /HDUQLQJ DQG 0RWLYDWLRQ 3UHOHF f 0DWFKLQJ PD[LPL]LQJ DQG WKH K\SHUEROLF UHLQIRUFHPHQW IHHGEDFN IXQFWLRQ 3V\FKRORJLFDO 5HYLHZ 5DFKOLQ + t *UHHQ / f &RPPLWPHQW FKRLFH DQG VHOIFRQWURO -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 5HHG 3 t 6FKDFKWPDQ 7 5 f ,QVWUXPHQWDO SHUIRUPDQFH RQ QHJDWLYH VFKHGXOHV 7KH 4XDUWHUO\ -RXUQDO RI ([SHULPHQWDO 3V\FKRORJ\ E 5HHG 3 t 6FKDFKWPDQ 7 5 f ,QVWUXPHQWDO UHVSRQGLQJ E\ UDWV RQ IUHH RSHUDQW FRQWLQJHQFLHV ZLWK FRPSRQHQWV WKDW VFKHGXOH UHVSRQVHGHSHQGHQW UHLQIRUFHU RPLVVLRQ $QLPDO /HDUQLQJ DQG %HKDYLRU 9
5RVHQIDUE 6 1HZODQG 0 & %UDQQRQ 6 ( t +RZH\ 6 f (IIHFWV RI VHOIJHQHUDWHG UXOHV RQ WKH GHYHORSPHQW RI VFKHGXOHFRQWUROOHG EHKDYLRU -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 6DYDVWDQR + t )DQWLQR ( ) f +XPDQ FKRLFH LQ FRQFXUUHQW UDWLRLQWHUYDO VFKHGXOHV RI UHLQIRUFHPHQW -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 6FKPLWW '5 f (IIHFWV RI UHLQIRUFHPHQW UDWH DQG UHLQIRUFHU PDJQLWXGH RQ FKRLFH EHKDYLRU RI KXPDQV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 6FKURHGHU 65 t +ROODQG -* f 5HLQIRUFHPHQW RI H\H PRYHPHQW ZLWK FRQFXUUHQW VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 6KLPRII ( 0DWWKHZV %$ t &DWDQLD $& f +XPDQ RSHUDQW SHUIRUPDQFH 6HQVLWLYLW\ DQG SVHXGRVHQVLWLYLW\ WR FRQWLQJHQFLHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 6LOEHUEHUJ $ 7KRPDV 5 t %HUHQG]HQ 1 f +XPDQ FKRLFH RQ FRQFXUUHQW YDULDEOHLQWHUYDO YDULDEOHUDWLR VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU 6NLQQHU % ) f 9HUEDO EHKDYLRU 1HZ
9DXJKDQ : -U t 0LOOHU + / f 2SWLPL]DWLRQ YHUVXV UHVSRQVH VWUHQJWK DFFRXQWV RI EHKDYLRU -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU :DQFKLVHQ % $ 7DWKDP 7 $ t +LQHOLQH 3 1 f 3LJHRQVn FKRLFHV LQ VLWXDWLRQV RI GLPLQLVKLQJ UHWXUQV )L[HGYHUVXV SURJUHVVLYHUDWLR VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU :DQFKLVHQ % $ 7DWKDP 7 $ t +LQHOLQH 3 1 f +XPDQ FKRLFH LQ FRXQWHULQWXLWLYH VLWXDWLRQV )L[HG YHUVXV SURJUHVVLYH VFKHGXOHV -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU :DVVHUPDQ ( $ t 1HXQDEHU f &ROOHJH VWXGHQWVn UHVSRQGLQJ WR DQG UDWLQJ RI FRQWLQJHQF\ UHODWLRQV 7KH UROH RI WHPSRUDO FRQWLJXLW\ -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU :HLQHU + f &RQWUROOLQJ KXPDQ IL[HGLQWHUYDO SHUIRUPDQFH -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU :HLQHU + f +XPDQ EHKDYLRUDO SHUVLVWHQFH 7KH 3V\FKRORJLFDO 5HFRUG :HLQHU + f 6RPH WKRXJKWV RQ GLVFUHSDQW KXPDQDQLPDO SHUIRUPDQFHV XQGHU VFKHGXOHV RI UHLQIRUFHPHQW 7KH 3V\FKRORJLFDO 5HFRUG =HLOHU 0' f 6FKHGXOHV RI UHLQIRUFHPHQW 7KH VOHHSLQJ JLDQW -RXUQDO RI WKH ([SHULPHQWDO $QDO\VLV RI %HKDYLRU
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